├── .DS_Store ├── .gradle ├── 6.1.1 │ ├── executionHistory │ │ ├── executionHistory.bin │ │ └── executionHistory.lock │ ├── fileChanges │ │ └── last-build.bin │ ├── fileContent │ │ └── fileContent.lock │ ├── fileHashes │ │ ├── fileHashes.bin │ │ ├── fileHashes.lock │ │ └── resourceHashesCache.bin │ ├── gc.properties │ └── javaCompile │ │ ├── classAnalysis.bin │ │ ├── jarAnalysis.bin │ │ ├── javaCompile.lock │ │ └── taskHistory.bin ├── buildOutputCleanup │ ├── buildOutputCleanup.lock │ ├── cache.properties │ └── outputFiles.bin ├── checksums │ ├── checksums.lock │ ├── md5-checksums.bin │ └── sha1-checksums.bin └── vcs-1 │ └── gc.properties ├── .idea ├── .gitignore ├── .name ├── caches │ └── build_file_checksums.ser ├── codeStyles │ ├── Project.xml │ └── codeStyleConfig.xml ├── gradle.xml ├── jarRepositories.xml ├── libraries │ ├── Gradle__androidx_activity_activity_1_0_0_aar.xml │ ├── Gradle__androidx_annotation_annotation_1_1_0_jar.xml │ ├── Gradle__androidx_appcompat_appcompat_1_1_0_aar.xml │ ├── Gradle__androidx_appcompat_appcompat_resources_1_1_0_aar.xml │ ├── Gradle__androidx_arch_core_core_common_2_1_0_jar.xml │ ├── Gradle__androidx_arch_core_core_runtime_2_0_0_aar.xml │ ├── Gradle__androidx_collection_collection_1_1_0_jar.xml │ ├── Gradle__androidx_constraintlayout_constraintlayout_1_1_3_aar.xml │ ├── Gradle__androidx_constraintlayout_constraintlayout_solver_1_1_3_jar.xml │ ├── Gradle__androidx_core_core_1_3_0_aar.xml │ ├── Gradle__androidx_core_core_ktx_1_3_0_aar.xml │ ├── Gradle__androidx_cursoradapter_cursoradapter_1_0_0_aar.xml │ ├── Gradle__androidx_customview_customview_1_0_0_aar.xml │ ├── Gradle__androidx_drawerlayout_drawerlayout_1_0_0_aar.xml │ ├── Gradle__androidx_fragment_fragment_1_1_0_aar.xml │ ├── Gradle__androidx_interpolator_interpolator_1_0_0_aar.xml │ ├── Gradle__androidx_lifecycle_lifecycle_common_2_1_0_jar.xml │ ├── Gradle__androidx_lifecycle_lifecycle_livedata_2_0_0_aar.xml │ ├── Gradle__androidx_lifecycle_lifecycle_livedata_core_2_0_0_aar.xml │ ├── Gradle__androidx_lifecycle_lifecycle_runtime_2_1_0_aar.xml │ ├── Gradle__androidx_lifecycle_lifecycle_viewmodel_2_1_0_aar.xml │ ├── Gradle__androidx_loader_loader_1_0_0_aar.xml │ ├── Gradle__androidx_savedstate_savedstate_1_0_0_aar.xml │ ├── Gradle__androidx_test_core_1_2_0_aar.xml │ ├── Gradle__androidx_test_espresso_espresso_core_3_2_0_aar.xml │ ├── Gradle__androidx_test_espresso_espresso_idling_resource_3_2_0_aar.xml │ ├── Gradle__androidx_test_ext_junit_1_1_1_aar.xml │ ├── Gradle__androidx_test_monitor_1_2_0_aar.xml │ ├── Gradle__androidx_test_runner_1_2_0_aar.xml │ ├── Gradle__androidx_vectordrawable_vectordrawable_1_1_0_aar.xml │ ├── Gradle__androidx_vectordrawable_vectordrawable_animated_1_1_0_aar.xml │ ├── Gradle__androidx_versionedparcelable_versionedparcelable_1_1_0_aar.xml │ ├── Gradle__androidx_viewpager_viewpager_1_0_0_aar.xml │ ├── Gradle__com_google_code_findbugs_jsr305_2_0_1_jar.xml │ ├── Gradle__com_squareup_javawriter_2_1_1_jar.xml │ ├── Gradle__javax_inject_javax_inject_1_jar.xml │ ├── Gradle__junit_junit_4_12_jar.xml │ ├── Gradle__net_sf_kxml_kxml2_2_3_0_jar.xml │ ├── Gradle__org_hamcrest_hamcrest_core_1_3_jar.xml │ ├── Gradle__org_hamcrest_hamcrest_integration_1_3_jar.xml │ ├── Gradle__org_hamcrest_hamcrest_library_1_3_jar.xml │ ├── Gradle__org_jetbrains_annotations_13_0_jar.xml │ ├── Gradle__org_jetbrains_kotlin_kotlin_android_extensions_runtime_1_3_72_jar.xml │ ├── Gradle__org_jetbrains_kotlin_kotlin_stdlib_1_3_72_jar.xml │ ├── Gradle__org_jetbrains_kotlin_kotlin_stdlib_common_1_3_72_jar.xml │ ├── Gradle__org_tensorflow_tensorflow_lite_0_0_0_nightly_aar.xml │ └── Gradle__org_tensorflow_tensorflow_lite_support_0_0_0_nightly_aar.xml ├── misc.xml ├── modules.xml ├── modules │ ├── MFCCCalculator.iml │ ├── NoiseClassifier_Android_Tensorflow.iml │ └── app │ │ └── app.iml ├── runConfigurations.xml └── vcs.xml ├── Android_NoiseClassification_Demo.mov ├── Android_NoiseClassification_demo.gif ├── Python_Notebook_Files ├── UrbanSound_200.csv ├── Urban_cnn_model.ipynb ├── Urban_data_preprocess.ipynb ├── model.tflite ├── test_data.csv ├── test_labels.csv ├── train_data.csv └── train_labels.csv ├── Readme.md ├── SampleAudio ├── 001_children_playing.wav ├── 001_dog_barking.wav ├── 001_street_music.wav └── 002_dog_barking.wav ├── app ├── .DS_Store ├── .gitignore ├── build.gradle ├── proguard-rules.pro └── src │ ├── .DS_Store │ ├── androidTest │ └── java │ │ └── com │ │ └── tensorflow │ │ └── android │ │ └── noiseclassifier │ │ └── ExampleInstrumentedTest.kt │ ├── main │ ├── .DS_Store │ ├── AndroidManifest.xml │ ├── assets │ │ ├── labels.txt │ │ └── model.tflite │ ├── java │ │ ├── .DS_Store │ │ └── com │ │ │ ├── .DS_Store │ │ │ └── tensorflow │ │ │ ├── .DS_Store │ │ │ └── android │ │ │ ├── .DS_Store │ │ │ ├── audio │ │ │ └── features │ │ │ │ ├── FFT.java │ │ │ │ ├── MFCC.java │ │ │ │ ├── WavFile.java │ │ │ │ └── WavFileException.java │ │ │ └── noiseclassifier │ │ │ ├── MainActivity.kt │ │ │ └── Recognition.kt │ └── res │ │ ├── drawable-v24 │ │ └── ic_launcher_foreground.xml │ │ ├── drawable │ │ └── ic_launcher_background.xml │ │ ├── layout │ │ └── activity_main.xml │ │ ├── mipmap-anydpi-v26 │ │ ├── ic_launcher.xml │ │ └── ic_launcher_round.xml │ │ ├── mipmap-hdpi │ │ ├── ic_launcher.png │ │ └── ic_launcher_round.png │ │ ├── mipmap-mdpi │ │ ├── ic_launcher.png │ │ └── ic_launcher_round.png │ │ ├── mipmap-xhdpi │ │ ├── ic_launcher.png │ │ └── ic_launcher_round.png │ │ ├── mipmap-xxhdpi │ │ ├── ic_launcher.png │ │ └── ic_launcher_round.png │ │ ├── mipmap-xxxhdpi │ │ ├── ic_launcher.png │ │ └── ic_launcher_round.png │ │ └── values │ │ ├── colors.xml │ │ ├── strings.xml │ │ └── styles.xml │ └── test │ └── java │ └── com │ └── tensorflow │ └── android │ └── noiseclassifier │ └── ExampleUnitTest.kt ├── build.gradle ├── gradle.properties ├── gradle └── wrapper │ ├── gradle-wrapper.jar │ └── gradle-wrapper.properties ├── gradlew ├── gradlew.bat ├── local.properties └── settings.gradle /.DS_Store: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.DS_Store -------------------------------------------------------------------------------- /.gradle/6.1.1/executionHistory/executionHistory.bin: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/6.1.1/executionHistory/executionHistory.bin -------------------------------------------------------------------------------- /.gradle/6.1.1/executionHistory/executionHistory.lock: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/6.1.1/executionHistory/executionHistory.lock -------------------------------------------------------------------------------- /.gradle/6.1.1/fileChanges/last-build.bin: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /.gradle/6.1.1/fileContent/fileContent.lock: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/6.1.1/fileContent/fileContent.lock -------------------------------------------------------------------------------- /.gradle/6.1.1/fileHashes/fileHashes.bin: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/6.1.1/fileHashes/fileHashes.bin -------------------------------------------------------------------------------- /.gradle/6.1.1/fileHashes/fileHashes.lock: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/6.1.1/fileHashes/fileHashes.lock -------------------------------------------------------------------------------- /.gradle/6.1.1/fileHashes/resourceHashesCache.bin: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/6.1.1/fileHashes/resourceHashesCache.bin -------------------------------------------------------------------------------- /.gradle/6.1.1/gc.properties: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/6.1.1/gc.properties -------------------------------------------------------------------------------- /.gradle/6.1.1/javaCompile/classAnalysis.bin: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/6.1.1/javaCompile/classAnalysis.bin -------------------------------------------------------------------------------- /.gradle/6.1.1/javaCompile/jarAnalysis.bin: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/6.1.1/javaCompile/jarAnalysis.bin -------------------------------------------------------------------------------- /.gradle/6.1.1/javaCompile/javaCompile.lock: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/6.1.1/javaCompile/javaCompile.lock -------------------------------------------------------------------------------- /.gradle/6.1.1/javaCompile/taskHistory.bin: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/6.1.1/javaCompile/taskHistory.bin -------------------------------------------------------------------------------- /.gradle/buildOutputCleanup/buildOutputCleanup.lock: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/buildOutputCleanup/buildOutputCleanup.lock -------------------------------------------------------------------------------- /.gradle/buildOutputCleanup/cache.properties: -------------------------------------------------------------------------------- 1 | #Mon Jun 22 14:54:09 CDT 2020 2 | gradle.version=6.1.1 3 | -------------------------------------------------------------------------------- /.gradle/buildOutputCleanup/outputFiles.bin: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/buildOutputCleanup/outputFiles.bin -------------------------------------------------------------------------------- /.gradle/checksums/checksums.lock: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/checksums/checksums.lock -------------------------------------------------------------------------------- /.gradle/checksums/md5-checksums.bin: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/checksums/md5-checksums.bin -------------------------------------------------------------------------------- /.gradle/checksums/sha1-checksums.bin: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/checksums/sha1-checksums.bin -------------------------------------------------------------------------------- /.gradle/vcs-1/gc.properties: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.gradle/vcs-1/gc.properties -------------------------------------------------------------------------------- /.idea/.gitignore: -------------------------------------------------------------------------------- 1 | # Default ignored files 2 | /workspace.xml -------------------------------------------------------------------------------- /.idea/.name: -------------------------------------------------------------------------------- 1 | MFCCCalculator -------------------------------------------------------------------------------- /.idea/caches/build_file_checksums.ser: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/.idea/caches/build_file_checksums.ser -------------------------------------------------------------------------------- /.idea/codeStyles/Project.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 7 | 8 | 9 |
10 | 11 | 12 | 13 | xmlns:android 14 | 15 | ^$ 16 | 17 | 18 | 19 |
20 |
21 | 22 | 23 | 24 | xmlns:.* 25 | 26 | ^$ 27 | 28 | 29 | BY_NAME 30 | 31 |
32 |
33 | 34 | 35 | 36 | .*:id 37 | 38 | http://schemas.android.com/apk/res/android 39 | 40 | 41 | 42 |
43 |
44 | 45 | 46 | 47 | .*:name 48 | 49 | http://schemas.android.com/apk/res/android 50 | 51 | 52 | 53 |
54 |
55 | 56 | 57 | 58 | name 59 | 60 | ^$ 61 | 62 | 63 | 64 |
65 |
66 | 67 | 68 | 69 | style 70 | 71 | ^$ 72 | 73 | 74 | 75 |
76 |
77 | 78 | 79 | 80 | .* 81 | 82 | ^$ 83 | 84 | 85 | BY_NAME 86 | 87 |
88 |
89 | 90 | 91 | 92 | .* 93 | 94 | http://schemas.android.com/apk/res/android 95 | 96 | 97 | ANDROID_ATTRIBUTE_ORDER 98 | 99 |
100 |
101 | 102 | 103 | 104 | .* 105 | 106 | .* 107 | 108 | 109 | BY_NAME 110 | 111 |
112 | 113 | 114 | 115 | 116 | -------------------------------------------------------------------------------- /.idea/codeStyles/codeStyleConfig.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 5 | -------------------------------------------------------------------------------- /.idea/gradle.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 20 | 21 | -------------------------------------------------------------------------------- /.idea/jarRepositories.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 9 | 10 | 14 | 15 | 19 | 20 | 24 | 25 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_activity_activity_1_0_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_annotation_annotation_1_1_0_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_appcompat_appcompat_1_1_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_appcompat_appcompat_resources_1_1_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_arch_core_core_common_2_1_0_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_arch_core_core_runtime_2_0_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_collection_collection_1_1_0_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_constraintlayout_constraintlayout_1_1_3_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_constraintlayout_constraintlayout_solver_1_1_3_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_core_core_1_3_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_core_core_ktx_1_3_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_cursoradapter_cursoradapter_1_0_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_customview_customview_1_0_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_drawerlayout_drawerlayout_1_0_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_fragment_fragment_1_1_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_interpolator_interpolator_1_0_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_lifecycle_lifecycle_common_2_1_0_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_lifecycle_lifecycle_livedata_2_0_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_lifecycle_lifecycle_livedata_core_2_0_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_lifecycle_lifecycle_runtime_2_1_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_lifecycle_lifecycle_viewmodel_2_1_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_loader_loader_1_0_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_savedstate_savedstate_1_0_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_test_core_1_2_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_test_espresso_espresso_core_3_2_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_test_espresso_espresso_idling_resource_3_2_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_test_ext_junit_1_1_1_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_test_monitor_1_2_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_test_runner_1_2_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_vectordrawable_vectordrawable_1_1_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_vectordrawable_vectordrawable_animated_1_1_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_versionedparcelable_versionedparcelable_1_1_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__androidx_viewpager_viewpager_1_0_0_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__com_google_code_findbugs_jsr305_2_0_1_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__com_squareup_javawriter_2_1_1_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__javax_inject_javax_inject_1_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__junit_junit_4_12_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__net_sf_kxml_kxml2_2_3_0_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__org_hamcrest_hamcrest_core_1_3_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__org_hamcrest_hamcrest_integration_1_3_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__org_hamcrest_hamcrest_library_1_3_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__org_jetbrains_annotations_13_0_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__org_jetbrains_kotlin_kotlin_android_extensions_runtime_1_3_72_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__org_jetbrains_kotlin_kotlin_stdlib_1_3_72_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__org_jetbrains_kotlin_kotlin_stdlib_common_1_3_72_jar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__org_tensorflow_tensorflow_lite_0_0_0_nightly_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | -------------------------------------------------------------------------------- /.idea/libraries/Gradle__org_tensorflow_tensorflow_lite_support_0_0_0_nightly_aar.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | -------------------------------------------------------------------------------- /.idea/misc.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 9 | -------------------------------------------------------------------------------- /.idea/modules.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | -------------------------------------------------------------------------------- /.idea/modules/MFCCCalculator.iml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | -------------------------------------------------------------------------------- /.idea/modules/NoiseClassifier_Android_Tensorflow.iml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | -------------------------------------------------------------------------------- /.idea/modules/app/app.iml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 10 | 11 | 12 | 13 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 | 46 | 47 | 48 | 49 | 50 | 51 | 52 | 53 | 54 | 55 | 56 | 57 | 58 | 59 | 60 | 61 | 62 | 63 | 64 | 65 | 66 | 67 | 68 | 69 | 70 | 71 | 72 | 73 | 74 | 75 | 76 | 77 | 78 | 79 | 80 | 81 | 82 | 83 | 84 | 85 | 86 | 87 | 88 | 89 | 90 | 91 | 92 | 93 | 94 | 95 | 96 | 97 | 98 | 99 | 100 | 101 | 102 | 103 | 104 | 105 | 106 | 107 | 108 | 109 | 110 | 111 | 112 | 113 | 114 | 115 | 116 | 117 | 118 | 119 | 120 | 121 | 122 | 123 | 124 | 125 | 126 | 127 | 128 | 129 | 130 | 131 | 132 | 133 | -------------------------------------------------------------------------------- /.idea/runConfigurations.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 11 | 12 | -------------------------------------------------------------------------------- /.idea/vcs.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | -------------------------------------------------------------------------------- /Android_NoiseClassification_Demo.mov: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/Android_NoiseClassification_Demo.mov -------------------------------------------------------------------------------- /Android_NoiseClassification_demo.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/Android_NoiseClassification_demo.gif -------------------------------------------------------------------------------- /Python_Notebook_Files/UrbanSound_200.csv: -------------------------------------------------------------------------------- 1 | slice_file_name,fsID,start,end,salience,fold,classID,class 2 | 100032-3-0-0.wav,100032,0.0,0.317551,1,5,3,dog_bark 3 | 100263-2-0-117.wav,100263,58.5,62.5,1,5,2,children_playing 4 | 100263-2-0-121.wav,100263,60.5,64.5,1,5,2,children_playing 5 | 100263-2-0-126.wav,100263,63.0,67.0,1,5,2,children_playing 6 | 100263-2-0-137.wav,100263,68.5,72.5,1,5,2,children_playing 7 | 100263-2-0-143.wav,100263,71.5,75.5,1,5,2,children_playing 8 | 100263-2-0-161.wav,100263,80.5,84.5,1,5,2,children_playing 9 | 100263-2-0-3.wav,100263,1.5,5.5,1,5,2,children_playing 10 | 100263-2-0-36.wav,100263,18.0,22.0,1,5,2,children_playing 11 | 100648-1-0-0.wav,100648,4.823402,5.471927,2,10,1,car_horn 12 | 100648-1-1-0.wav,100648,8.998279,10.052132,2,10,1,car_horn 13 | 100648-1-2-0.wav,100648,16.699509,17.104837,2,10,1,car_horn 14 | 100648-1-3-0.wav,100648,17.631764,19.253075,2,10,1,car_horn 15 | 100648-1-4-0.wav,100648,25.332994,27.197502,2,10,1,car_horn 16 | 100652-3-0-0.wav,100652,0.0,4.0,1,2,3,dog_bark 17 | 100652-3-0-1.wav,100652,0.5,4.5,1,2,3,dog_bark 18 | 100652-3-0-2.wav,100652,1.0,5.0,1,2,3,dog_bark 19 | 100652-3-0-3.wav,100652,1.5,5.5,1,2,3,dog_bark 20 | 100795-3-0-0.wav,100795,0.19179,4.19179,1,10,3,dog_bark 21 | 100795-3-1-0.wav,100795,13.059155,17.059155,1,10,3,dog_bark 22 | 100795-3-1-1.wav,100795,13.559155,17.559155,1,10,3,dog_bark 23 | 100795-3-1-2.wav,100795,14.059155,18.059155,1,10,3,dog_bark 24 | 100852-0-0-0.wav,100852,0.0,4.0,1,5,0,air_conditioner 25 | 100852-0-0-1.wav,100852,0.5,4.5,1,5,0,air_conditioner 26 | 100852-0-0-10.wav,100852,5.0,9.0,1,5,0,air_conditioner 27 | 100852-0-0-11.wav,100852,5.5,9.5,1,5,0,air_conditioner 28 | 100852-0-0-12.wav,100852,6.0,10.0,1,5,0,air_conditioner 29 | 100852-0-0-13.wav,100852,6.5,10.5,1,5,0,air_conditioner 30 | 100852-0-0-14.wav,100852,7.0,11.0,1,5,0,air_conditioner 31 | 100852-0-0-15.wav,100852,7.5,11.5,1,5,0,air_conditioner 32 | 100852-0-0-16.wav,100852,8.0,12.0,1,5,0,air_conditioner 33 | 100852-0-0-17.wav,100852,8.5,12.5,1,5,0,air_conditioner 34 | 100852-0-0-18.wav,100852,9.0,13.0,1,5,0,air_conditioner 35 | 100852-0-0-19.wav,100852,9.5,13.5,1,5,0,air_conditioner 36 | 100852-0-0-2.wav,100852,1.0,5.0,1,5,0,air_conditioner 37 | 100852-0-0-20.wav,100852,10.0,14.0,1,5,0,air_conditioner 38 | 100852-0-0-21.wav,100852,10.5,14.5,1,5,0,air_conditioner 39 | 100852-0-0-22.wav,100852,11.0,15.0,1,5,0,air_conditioner 40 | 100852-0-0-23.wav,100852,11.5,15.5,1,5,0,air_conditioner 41 | 100852-0-0-24.wav,100852,12.0,16.0,1,5,0,air_conditioner 42 | 100852-0-0-25.wav,100852,12.5,16.5,1,5,0,air_conditioner 43 | 100852-0-0-26.wav,100852,13.0,17.0,1,5,0,air_conditioner 44 | 100852-0-0-27.wav,100852,13.5,17.5,1,5,0,air_conditioner 45 | 100852-0-0-28.wav,100852,14.0,18.0,1,5,0,air_conditioner 46 | 100852-0-0-29.wav,100852,14.5,18.5,1,5,0,air_conditioner 47 | 100852-0-0-3.wav,100852,1.5,5.5,1,5,0,air_conditioner 48 | 100852-0-0-30.wav,100852,15.0,19.0,1,5,0,air_conditioner 49 | 100852-0-0-4.wav,100852,2.0,6.0,1,5,0,air_conditioner 50 | 100852-0-0-5.wav,100852,2.5,6.5,1,5,0,air_conditioner 51 | 100852-0-0-6.wav,100852,3.0,7.0,1,5,0,air_conditioner 52 | 100852-0-0-7.wav,100852,3.5,7.5,1,5,0,air_conditioner 53 | 100852-0-0-8.wav,100852,4.0,8.0,1,5,0,air_conditioner 54 | 100852-0-0-9.wav,100852,4.5,8.5,1,5,0,air_conditioner 55 | 101281-3-0-0.wav,101281,0.341333,4.341333,2,6,3,dog_bark 56 | 101281-3-0-14.wav,101281,7.341333,11.341333000000000,2,6,3,dog_bark 57 | 101281-3-0-5.wav,101281,2.841333,6.8413330000000000,2,6,3,dog_bark 58 | 101382-2-0-10.wav,101382,5.0,9.0,1,10,2,children_playing 59 | 101382-2-0-12.wav,101382,6.0,10.0,1,10,2,children_playing 60 | 101382-2-0-20.wav,101382,10.0,14.0,1,10,2,children_playing 61 | 101382-2-0-21.wav,101382,10.5,14.5,1,10,2,children_playing 62 | 101382-2-0-29.wav,101382,14.5,18.5,1,10,2,children_playing 63 | 101382-2-0-33.wav,101382,16.5,20.5,1,10,2,children_playing 64 | 101382-2-0-42.wav,101382,21.0,25.0,1,10,2,children_playing 65 | 101382-2-0-45.wav,101382,22.5,26.5,1,10,2,children_playing 66 | 101415-3-0-2.wav,101415,1.0,5.0,1,1,3,dog_bark 67 | 101415-3-0-3.wav,101415,1.5,5.5,1,1,3,dog_bark 68 | 101415-3-0-8.wav,101415,4.0,8.0,1,1,3,dog_bark 69 | 101729-0-0-1.wav,101729,0.5,4.5,2,9,0,air_conditioner 70 | 101729-0-0-11.wav,101729,5.5,9.5,2,9,0,air_conditioner 71 | 101729-0-0-12.wav,101729,6.0,10.0,2,9,0,air_conditioner 72 | 101729-0-0-13.wav,101729,6.5,10.5,2,9,0,air_conditioner 73 | 101729-0-0-14.wav,101729,7.0,11.0,2,9,0,air_conditioner 74 | 101729-0-0-16.wav,101729,8.0,12.0,2,9,0,air_conditioner 75 | 101729-0-0-17.wav,101729,8.5,12.5,2,9,0,air_conditioner 76 | 101729-0-0-18.wav,101729,9.0,13.0,2,9,0,air_conditioner 77 | 101729-0-0-19.wav,101729,9.5,13.5,2,9,0,air_conditioner 78 | 101729-0-0-21.wav,101729,10.5,14.5,2,9,0,air_conditioner 79 | 101729-0-0-22.wav,101729,11.0,15.0,2,9,0,air_conditioner 80 | 101729-0-0-23.wav,101729,11.5,15.5,2,9,0,air_conditioner 81 | 101729-0-0-24.wav,101729,12.0,16.0,2,9,0,air_conditioner 82 | 101729-0-0-26.wav,101729,13.0,17.0,2,9,0,air_conditioner 83 | 101729-0-0-28.wav,101729,14.0,18.0,2,9,0,air_conditioner 84 | 101729-0-0-29.wav,101729,14.5,18.5,2,9,0,air_conditioner 85 | 101729-0-0-3.wav,101729,1.5,5.5,2,9,0,air_conditioner 86 | 101729-0-0-32.wav,101729,16.0,20.0,2,9,0,air_conditioner 87 | 101729-0-0-33.wav,101729,16.5,20.5,2,9,0,air_conditioner 88 | 101729-0-0-36.wav,101729,18.0,22.0,2,9,0,air_conditioner 89 | 101729-0-0-37.wav,101729,18.5,22.5,2,9,0,air_conditioner 90 | 101729-0-0-38.wav,101729,19.0,23.0,2,9,0,air_conditioner 91 | 101729-0-0-39.wav,101729,19.5,23.5,2,9,0,air_conditioner 92 | 101729-0-0-4.wav,101729,2.0,6.0,2,9,0,air_conditioner 93 | 101729-0-0-40.wav,101729,20.0,24.0,2,9,0,air_conditioner 94 | 101729-0-0-6.wav,101729,3.0,7.0,2,9,0,air_conditioner 95 | 101729-0-0-9.wav,101729,4.5,8.5,2,9,0,air_conditioner 96 | 101848-9-0-0.wav,101848,0.0,4.0,1,7,9,street_music 97 | 101848-9-0-1.wav,101848,0.5,4.5,1,7,9,street_music 98 | 101848-9-0-2.wav,101848,1.0,5.0,1,7,9,street_music 99 | 101848-9-0-3.wav,101848,1.5,5.5,1,7,9,street_music 100 | 101848-9-0-8.wav,101848,4.0,8.0,1,7,9,street_music 101 | 101848-9-0-9.wav,101848,4.5,8.5,1,7,9,street_music 102 | 102102-3-0-0.wav,102102,1.573952,4.673824,2,4,3,dog_bark 103 | 102103-3-0-0.wav,102103,1.97603,2.508892,2,10,3,dog_bark 104 | 102103-3-1-0.wav,102103,4.179637,4.568182,2,10,3,dog_bark 105 | 102104-3-0-0.wav,102104,2.429279,3.241121,2,2,3,dog_bark 106 | 102105-3-0-0.wav,102105,2.041889,6.041889000000000,2,3,3,dog_bark 107 | 102106-3-0-0.wav,102106,2.243852,3.884477,2,1,3,dog_bark 108 | 102305-6-0-0.wav,102305,0.0,2.61161,1,1,6,gun_shot 109 | 102547-3-0-2.wav,102547,1.0,5.0,1,5,3,dog_bark 110 | 102547-3-0-7.wav,102547,3.5,7.5,1,5,3,dog_bark 111 | 102547-3-0-8.wav,102547,4.0,8.0,1,5,3,dog_bark 112 | 102842-3-0-1.wav,102842,0.5,4.5,2,1,3,dog_bark 113 | 102842-3-1-0.wav,102842,113.061785,117.061785,2,1,3,dog_bark 114 | 102842-3-1-5.wav,102842,115.561785,119.561785,2,1,3,dog_bark 115 | 102842-3-1-6.wav,102842,116.061785,120.061785,2,1,3,dog_bark 116 | 102853-8-0-0.wav,102853,0.0,4.0,2,7,8,siren 117 | 102853-8-0-1.wav,102853,0.5,4.5,2,7,8,siren 118 | 102853-8-0-2.wav,102853,1.0,5.0,2,7,8,siren 119 | 102853-8-0-3.wav,102853,1.5,5.5,2,7,8,siren 120 | 102853-8-0-4.wav,102853,2.0,6.0,2,7,8,siren 121 | 102853-8-0-5.wav,102853,2.5,6.5,2,7,8,siren 122 | 102853-8-0-6.wav,102853,3.0,7.0,2,7,8,siren 123 | 102853-8-1-0.wav,102853,27.214044,31.214044,2,7,8,siren 124 | 102857-5-0-0.wav,102857,0.0,4.0,1,10,5,engine_idling 125 | 102857-5-0-1.wav,102857,0.5,4.5,1,10,5,engine_idling 126 | 102857-5-0-10.wav,102857,5.0,9.0,1,10,5,engine_idling 127 | 102857-5-0-11.wav,102857,5.5,9.5,1,10,5,engine_idling 128 | 102857-5-0-12.wav,102857,6.0,10.0,1,10,5,engine_idling 129 | 102857-5-0-13.wav,102857,6.5,10.5,1,10,5,engine_idling 130 | 102857-5-0-14.wav,102857,7.0,11.0,1,10,5,engine_idling 131 | 102857-5-0-15.wav,102857,7.5,11.5,1,10,5,engine_idling 132 | 102857-5-0-16.wav,102857,8.0,12.0,1,10,5,engine_idling 133 | 102857-5-0-17.wav,102857,8.5,12.5,1,10,5,engine_idling 134 | 102857-5-0-18.wav,102857,9.0,13.0,1,10,5,engine_idling 135 | 102857-5-0-19.wav,102857,9.5,13.5,1,10,5,engine_idling 136 | 102857-5-0-2.wav,102857,1.0,5.0,1,10,5,engine_idling 137 | 102857-5-0-20.wav,102857,10.0,14.0,1,10,5,engine_idling 138 | 102857-5-0-21.wav,102857,10.5,14.5,1,10,5,engine_idling 139 | 102857-5-0-22.wav,102857,11.0,15.0,1,10,5,engine_idling 140 | 102857-5-0-23.wav,102857,11.5,15.5,1,10,5,engine_idling 141 | 102857-5-0-24.wav,102857,12.0,16.0,1,10,5,engine_idling 142 | 102857-5-0-25.wav,102857,12.5,16.5,1,10,5,engine_idling 143 | 102857-5-0-26.wav,102857,13.0,17.0,1,10,5,engine_idling 144 | 102857-5-0-27.wav,102857,13.5,17.5,1,10,5,engine_idling 145 | 102857-5-0-28.wav,102857,14.0,18.0,1,10,5,engine_idling 146 | 102857-5-0-29.wav,102857,14.5,18.5,1,10,5,engine_idling 147 | 102857-5-0-3.wav,102857,1.5,5.5,1,10,5,engine_idling 148 | 102857-5-0-30.wav,102857,15.0,19.0,1,10,5,engine_idling 149 | 102857-5-0-4.wav,102857,2.0,6.0,1,10,5,engine_idling 150 | 102857-5-0-5.wav,102857,2.5,6.5,1,10,5,engine_idling 151 | 102857-5-0-6.wav,102857,3.0,7.0,1,10,5,engine_idling 152 | 102857-5-0-7.wav,102857,3.5,7.5,1,10,5,engine_idling 153 | 102857-5-0-8.wav,102857,4.0,8.0,1,10,5,engine_idling 154 | 102857-5-0-9.wav,102857,4.5,8.5,1,10,5,engine_idling 155 | 102858-5-0-0.wav,102858,0.0,4.0,1,2,5,engine_idling 156 | 102858-5-0-1.wav,102858,0.5,4.5,1,2,5,engine_idling 157 | 102871-8-0-0.wav,102871,0.0,4.0,1,2,8,siren 158 | 102871-8-0-1.wav,102871,0.5,4.5,1,2,8,siren 159 | 102871-8-0-10.wav,102871,5.0,9.0,1,2,8,siren 160 | 102871-8-0-11.wav,102871,5.5,9.5,1,2,8,siren 161 | 102871-8-0-12.wav,102871,6.0,10.0,1,2,8,siren 162 | 102871-8-0-13.wav,102871,6.5,10.5,1,2,8,siren 163 | 102871-8-0-14.wav,102871,7.0,11.0,1,2,8,siren 164 | 102871-8-0-15.wav,102871,7.5,11.5,1,2,8,siren 165 | 102871-8-0-2.wav,102871,1.0,5.0,1,2,8,siren 166 | 102871-8-0-3.wav,102871,1.5,5.5,1,2,8,siren 167 | 102871-8-0-4.wav,102871,2.0,6.0,1,2,8,siren 168 | 102871-8-0-5.wav,102871,2.5,6.5,1,2,8,siren 169 | 102871-8-0-6.wav,102871,3.0,7.0,1,2,8,siren 170 | 102871-8-0-7.wav,102871,3.5,7.5,1,2,8,siren 171 | 102871-8-0-8.wav,102871,4.0,8.0,1,2,8,siren 172 | 102871-8-0-9.wav,102871,4.5,8.5,1,2,8,siren 173 | 103074-7-0-0.wav,103074,3.341279,7.341279,1,1,7,jackhammer 174 | 103074-7-0-1.wav,103074,3.841279,7.841279,1,1,7,jackhammer 175 | 103074-7-0-2.wav,103074,4.341279,8.341279,1,1,7,jackhammer 176 | 103074-7-1-0.wav,103074,22.517317,26.517317,1,1,7,jackhammer 177 | 103074-7-1-1.wav,103074,23.017317,27.017317,1,1,7,jackhammer 178 | 103074-7-1-2.wav,103074,23.517317,27.517317,1,1,7,jackhammer 179 | 103074-7-1-3.wav,103074,24.017317,28.017317,1,1,7,jackhammer 180 | 103074-7-1-4.wav,103074,24.517317,28.517317,1,1,7,jackhammer 181 | 103074-7-1-5.wav,103074,25.017317,29.017317,1,1,7,jackhammer 182 | 103074-7-1-6.wav,103074,25.517317,29.517317,1,1,7,jackhammer 183 | 103074-7-2-0.wav,103074,42.347083,43.727176,1,1,7,jackhammer 184 | 103074-7-3-0.wav,103074,46.632637,50.632637,1,1,7,jackhammer 185 | 103074-7-3-1.wav,103074,47.132637,51.132637,1,1,7,jackhammer 186 | 103074-7-3-2.wav,103074,47.632637,51.632637,1,1,7,jackhammer 187 | 103074-7-3-3.wav,103074,48.132637,52.132637,1,1,7,jackhammer 188 | 103074-7-4-0.wav,103074,60.651482,64.651482,1,1,7,jackhammer 189 | 103074-7-4-1.wav,103074,61.151482,65.151482,1,1,7,jackhammer 190 | 103074-7-4-2.wav,103074,61.651482,65.651482,1,1,7,jackhammer 191 | 103074-7-4-3.wav,103074,62.151482,66.151482,1,1,7,jackhammer 192 | 103074-7-4-4.wav,103074,62.651482,66.651482,1,1,7,jackhammer 193 | 103074-7-4-5.wav,103074,63.151482,67.151482,1,1,7,jackhammer 194 | 103074-7-4-6.wav,103074,63.651482,67.651482,1,1,7,jackhammer 195 | 103076-3-0-0.wav,103076,0.0,2.527184,1,8,3,dog_bark 196 | 103076-3-1-0.wav,103076,6.788349,7.756796,1,8,3,dog_bark 197 | 103076-3-2-0.wav,103076,10.265534,10.717475,1,8,3,dog_bark 198 | 103199-4-0-0.wav,103199,0.0,4.0,1,3,4,drilling 199 | 103199-4-0-3.wav,103199,1.5,5.5,1,3,4,drilling 200 | 103199-4-0-4.wav,103199,2.0,6.0,1,3,4,drilling 201 | 103199-4-0-5.wav,103199,2.5,6.5,1,3,4,drilling 202 | 103199-4-0-6.wav,103199,3.0,7.0,1,3,4,drilling 203 | 103199-4-1-0.wav,103199,16.097136,17.617205,2,3,4,drilling 204 | 103199-4-2-0.wav,103199,17.617205,21.617205,1,3,4,drilling 205 | -------------------------------------------------------------------------------- /Python_Notebook_Files/Urban_cnn_model.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "code", 5 | "execution_count": 1, 6 | "metadata": { 7 | "colab": {}, 8 | "colab_type": "code", 9 | "id": "tA_zGJTRf_gC" 10 | }, 11 | "outputs": [], 12 | "source": [ 13 | "import numpy as np" 14 | ] 15 | }, 16 | { 17 | "cell_type": "code", 18 | "execution_count": 2, 19 | "metadata": { 20 | "colab": {}, 21 | "colab_type": "code", 22 | "id": "5uRr-bWZWzAD" 23 | }, 24 | "outputs": [], 25 | "source": [ 26 | "#extracting data from csv files into numpy arrays\n", 27 | "from numpy import genfromtxt\n", 28 | "x_train = genfromtxt('train_data.csv', delimiter=',')\n", 29 | "y_train = genfromtxt('train_labels.csv', delimiter=',')\n", 30 | "x_test = genfromtxt('test_data.csv', delimiter=',')\n", 31 | "y_test = genfromtxt('test_labels.csv', delimiter=',')" 32 | ] 33 | }, 34 | { 35 | "cell_type": "code", 36 | "execution_count": 3, 37 | "metadata": { 38 | "colab": { 39 | "base_uri": "https://localhost:8080/", 40 | "height": 34 41 | }, 42 | "colab_type": "code", 43 | "id": "sUE1btJWW3bw", 44 | "outputId": "f74bc8b1-fee5-4efd-b851-b2337f69afb9" 45 | }, 46 | "outputs": [ 47 | { 48 | "data": { 49 | "text/plain": [ 50 | "((153, 40), (50, 40), (153,), (50,))" 51 | ] 52 | }, 53 | "execution_count": 3, 54 | "metadata": {}, 55 | "output_type": "execute_result" 56 | } 57 | ], 58 | "source": [ 59 | "#shape\n", 60 | "x_train.shape,x_test.shape,y_train.shape,y_test.shape" 61 | ] 62 | }, 63 | { 64 | "cell_type": "code", 65 | "execution_count": 4, 66 | "metadata": { 67 | "colab": { 68 | "base_uri": "https://localhost:8080/", 69 | "height": 34 70 | }, 71 | "colab_type": "code", 72 | "id": "UDbtZ-GaXEXS", 73 | "outputId": "50df72f3-2b7e-45b7-ea23-80e5aaca1743" 74 | }, 75 | "outputs": [ 76 | { 77 | "data": { 78 | "text/plain": [ 79 | "((153, 10), (50, 10))" 80 | ] 81 | }, 82 | "execution_count": 4, 83 | "metadata": {}, 84 | "output_type": "execute_result" 85 | } 86 | ], 87 | "source": [ 88 | "#converting to one hot\n", 89 | "from tensorflow.keras.utils import to_categorical\n", 90 | "y_train = to_categorical(y_train, num_classes=10)\n", 91 | "y_test = to_categorical(y_test, num_classes=10)\n", 92 | "y_train.shape,y_test.shape" 93 | ] 94 | }, 95 | { 96 | "cell_type": "code", 97 | "execution_count": 6, 98 | "metadata": { 99 | "colab": {}, 100 | "colab_type": "code", 101 | "id": "15K2y9sUgn_L" 102 | }, 103 | "outputs": [], 104 | "source": [ 105 | "#reshaping to shape required by CNN\n", 106 | "x_train=np.reshape(x_train,(x_train.shape[0], 40,1,1))\n", 107 | "x_test=np.reshape(x_test,(x_test.shape[0], 40,1,1))" 108 | ] 109 | }, 110 | { 111 | "cell_type": "code", 112 | "execution_count": 7, 113 | "metadata": { 114 | "colab": { 115 | "base_uri": "https://localhost:8080/", 116 | "height": 34 117 | }, 118 | "colab_type": "code", 119 | "id": "CgeQAvcdgtwn", 120 | "outputId": "fb3d635b-1da0-4c3d-f17f-50c9619cf9da" 121 | }, 122 | "outputs": [ 123 | { 124 | "data": { 125 | "text/plain": [ 126 | "((153, 40, 1, 1), (50, 40, 1, 1))" 127 | ] 128 | }, 129 | "execution_count": 7, 130 | "metadata": {}, 131 | "output_type": "execute_result" 132 | } 133 | ], 134 | "source": [ 135 | "#shapes\n", 136 | "x_train.shape,x_test.shape" 137 | ] 138 | }, 139 | { 140 | "cell_type": "code", 141 | "execution_count": 8, 142 | "metadata": { 143 | "colab": {}, 144 | "colab_type": "code", 145 | "id": "3bQ2gHc8f9n4" 146 | }, 147 | "outputs": [ 148 | { 149 | "name": "stdout", 150 | "output_type": "stream", 151 | "text": [ 152 | "2.2.0\n" 153 | ] 154 | } 155 | ], 156 | "source": [ 157 | "from tensorflow.keras import Sequential\n", 158 | "import tensorflow as tf\n", 159 | "print(tf.__version__)\n", 160 | "from tensorflow.keras.layers import Dense, Flatten, Conv2D, MaxPooling2D, Dropout\n", 161 | "from tensorflow.keras import Model" 162 | ] 163 | }, 164 | { 165 | "cell_type": "code", 166 | "execution_count": 9, 167 | "metadata": { 168 | "colab": {}, 169 | "colab_type": "code", 170 | "id": "SyjNa0N0gGlY" 171 | }, 172 | "outputs": [], 173 | "source": [ 174 | "#forming model\n", 175 | "model=Sequential()" 176 | ] 177 | }, 178 | { 179 | "cell_type": "code", 180 | "execution_count": 10, 181 | "metadata": { 182 | "colab": {}, 183 | "colab_type": "code", 184 | "id": "fUJUiyVTgkGZ" 185 | }, 186 | "outputs": [], 187 | "source": [ 188 | "#adding layers and forming the model\n", 189 | "model.add(Conv2D(64,kernel_size=5,strides=1,padding=\"Same\",activation=\"relu\",input_shape=(40,1,1)))\n", 190 | "model.add(MaxPooling2D(padding=\"same\"))\n", 191 | "\n", 192 | "model.add(Conv2D(128,kernel_size=5,strides=1,padding=\"same\",activation=\"relu\"))\n", 193 | "model.add(MaxPooling2D(padding=\"same\"))\n", 194 | "model.add(Dropout(0.3))\n", 195 | "\n", 196 | "model.add(Flatten())\n", 197 | "\n", 198 | "model.add(Dense(256,activation=\"relu\"))\n", 199 | "model.add(Dropout(0.3))\n", 200 | "\n", 201 | "model.add(Dense(512,activation=\"relu\"))\n", 202 | "model.add(Dropout(0.3))\n", 203 | "\n", 204 | "model.add(Dense(10,activation=\"softmax\"))" 205 | ] 206 | }, 207 | { 208 | "cell_type": "code", 209 | "execution_count": 11, 210 | "metadata": { 211 | "colab": {}, 212 | "colab_type": "code", 213 | "id": "ZHek_bnng3xp" 214 | }, 215 | "outputs": [], 216 | "source": [ 217 | "#compiling\n", 218 | "model.compile(optimizer=\"adam\",loss=\"categorical_crossentropy\",metrics=[\"accuracy\"])" 219 | ] 220 | }, 221 | { 222 | "cell_type": "code", 223 | "execution_count": 12, 224 | "metadata": { 225 | "colab": { 226 | "base_uri": "https://localhost:8080/", 227 | "height": 1091 228 | }, 229 | "colab_type": "code", 230 | "id": "Hc_E0cTog96t", 231 | "outputId": "4fb76b2e-08bc-4a38-d574-79490df067f5" 232 | }, 233 | "outputs": [ 234 | { 235 | "name": "stdout", 236 | "output_type": "stream", 237 | "text": [ 238 | "Epoch 1/30\n", 239 | "4/4 [==============================] - 0s 48ms/step - loss: 2.0726 - accuracy: 0.2745 - val_loss: 3.3352 - val_accuracy: 0.1200\n", 240 | "Epoch 2/30\n", 241 | "4/4 [==============================] - 0s 13ms/step - loss: 1.6048 - accuracy: 0.4967 - val_loss: 3.6241 - val_accuracy: 0.1000\n", 242 | "Epoch 3/30\n", 243 | "4/4 [==============================] - 0s 13ms/step - loss: 1.3376 - accuracy: 0.5556 - val_loss: 4.1635 - val_accuracy: 0.1200\n", 244 | "Epoch 4/30\n", 245 | "4/4 [==============================] - 0s 13ms/step - loss: 1.0308 - accuracy: 0.7255 - val_loss: 4.1981 - val_accuracy: 0.1200\n", 246 | "Epoch 5/30\n", 247 | "4/4 [==============================] - 0s 13ms/step - loss: 0.7843 - accuracy: 0.7778 - val_loss: 4.3473 - val_accuracy: 0.1200\n", 248 | "Epoch 6/30\n", 249 | "4/4 [==============================] - 0s 12ms/step - loss: 0.7192 - accuracy: 0.7974 - val_loss: 4.4873 - val_accuracy: 0.1200\n", 250 | "Epoch 7/30\n", 251 | "4/4 [==============================] - 0s 13ms/step - loss: 0.5531 - accuracy: 0.8170 - val_loss: 4.0361 - val_accuracy: 0.1200\n", 252 | "Epoch 8/30\n", 253 | "4/4 [==============================] - 0s 13ms/step - loss: 0.4270 - accuracy: 0.8693 - val_loss: 3.6794 - val_accuracy: 0.1200\n", 254 | "Epoch 9/30\n", 255 | "4/4 [==============================] - 0s 13ms/step - loss: 0.3672 - accuracy: 0.9281 - val_loss: 3.8678 - val_accuracy: 0.1200\n", 256 | "Epoch 10/30\n", 257 | "4/4 [==============================] - 0s 14ms/step - loss: 0.2402 - accuracy: 0.9412 - val_loss: 4.3586 - val_accuracy: 0.1200\n", 258 | "Epoch 11/30\n", 259 | "4/4 [==============================] - 0s 12ms/step - loss: 0.2890 - accuracy: 0.9150 - val_loss: 3.7867 - val_accuracy: 0.1200\n", 260 | "Epoch 12/30\n", 261 | "4/4 [==============================] - 0s 12ms/step - loss: 0.1881 - accuracy: 0.9346 - val_loss: 3.5562 - val_accuracy: 0.1800\n", 262 | "Epoch 13/30\n", 263 | "4/4 [==============================] - 0s 12ms/step - loss: 0.2113 - accuracy: 0.9346 - val_loss: 3.8496 - val_accuracy: 0.7400\n", 264 | "Epoch 14/30\n", 265 | "4/4 [==============================] - 0s 12ms/step - loss: 0.2429 - accuracy: 0.9477 - val_loss: 3.5848 - val_accuracy: 0.7400\n", 266 | "Epoch 15/30\n", 267 | "4/4 [==============================] - 0s 13ms/step - loss: 0.1477 - accuracy: 0.9542 - val_loss: 3.2284 - val_accuracy: 0.7400\n", 268 | "Epoch 16/30\n", 269 | "4/4 [==============================] - 0s 15ms/step - loss: 0.1005 - accuracy: 0.9673 - val_loss: 3.2545 - val_accuracy: 0.7400\n", 270 | "Epoch 17/30\n", 271 | "4/4 [==============================] - 0s 16ms/step - loss: 0.0673 - accuracy: 0.9935 - val_loss: 3.5861 - val_accuracy: 0.7400\n", 272 | "Epoch 18/30\n", 273 | "4/4 [==============================] - 0s 16ms/step - loss: 0.0516 - accuracy: 0.9869 - val_loss: 3.7681 - val_accuracy: 0.7400\n", 274 | "Epoch 19/30\n", 275 | "4/4 [==============================] - 0s 15ms/step - loss: 0.0492 - accuracy: 0.9869 - val_loss: 4.0435 - val_accuracy: 0.7400\n", 276 | "Epoch 20/30\n", 277 | "4/4 [==============================] - 0s 17ms/step - loss: 0.0492 - accuracy: 0.9804 - val_loss: 4.2710 - val_accuracy: 0.7400\n", 278 | "Epoch 21/30\n", 279 | "4/4 [==============================] - 0s 15ms/step - loss: 0.0373 - accuracy: 0.9869 - val_loss: 4.4171 - val_accuracy: 0.7400\n", 280 | "Epoch 22/30\n", 281 | "4/4 [==============================] - 0s 15ms/step - loss: 0.0256 - accuracy: 0.9935 - val_loss: 4.4984 - val_accuracy: 0.7400\n", 282 | "Epoch 23/30\n", 283 | "4/4 [==============================] - 0s 15ms/step - loss: 0.0367 - accuracy: 0.9869 - val_loss: 4.3313 - val_accuracy: 0.7400\n", 284 | "Epoch 24/30\n", 285 | "4/4 [==============================] - 0s 15ms/step - loss: 0.0494 - accuracy: 0.9869 - val_loss: 4.3794 - val_accuracy: 0.7400\n", 286 | "Epoch 25/30\n", 287 | "4/4 [==============================] - 0s 15ms/step - loss: 0.0083 - accuracy: 1.0000 - val_loss: 4.4873 - val_accuracy: 0.7400\n", 288 | "Epoch 26/30\n", 289 | "4/4 [==============================] - 0s 16ms/step - loss: 0.0132 - accuracy: 0.9935 - val_loss: 4.6293 - val_accuracy: 0.7400\n", 290 | "Epoch 27/30\n", 291 | "4/4 [==============================] - 0s 15ms/step - loss: 0.0276 - accuracy: 0.9935 - val_loss: 4.4349 - val_accuracy: 0.7400\n", 292 | "Epoch 28/30\n", 293 | "4/4 [==============================] - 0s 17ms/step - loss: 0.0114 - accuracy: 1.0000 - val_loss: 4.4444 - val_accuracy: 0.7400\n", 294 | "Epoch 29/30\n", 295 | "4/4 [==============================] - 0s 14ms/step - loss: 0.0135 - accuracy: 1.0000 - val_loss: 4.6879 - val_accuracy: 0.7400\n", 296 | "Epoch 30/30\n", 297 | "4/4 [==============================] - 0s 15ms/step - loss: 0.0295 - accuracy: 0.9935 - val_loss: 4.8164 - val_accuracy: 0.7400\n" 298 | ] 299 | }, 300 | { 301 | "data": { 302 | "text/plain": [ 303 | "" 304 | ] 305 | }, 306 | "execution_count": 12, 307 | "metadata": {}, 308 | "output_type": "execute_result" 309 | } 310 | ], 311 | "source": [ 312 | "#training the model\n", 313 | "model.fit(x_train,y_train,batch_size=50,epochs=30,validation_data=(x_test,y_test))" 314 | ] 315 | }, 316 | { 317 | "cell_type": "code", 318 | "execution_count": 13, 319 | "metadata": { 320 | "colab": { 321 | "base_uri": "https://localhost:8080/", 322 | "height": 85 323 | }, 324 | "colab_type": "code", 325 | "id": "N1Rp9FTYhBpZ", 326 | "outputId": "8ac2e5be-314a-4284-ea31-ddbfeda1d4de" 327 | }, 328 | "outputs": [ 329 | { 330 | "name": "stdout", 331 | "output_type": "stream", 332 | "text": [ 333 | "5/5 [==============================] - 0s 2ms/step - loss: 0.0034 - accuracy: 1.0000\n", 334 | "2/2 [==============================] - 0s 2ms/step - loss: 4.8164 - accuracy: 0.7400\n", 335 | "[0.0033533996902406216, 1.0]\n", 336 | "[4.816380023956299, 0.7400000095367432]\n" 337 | ] 338 | } 339 | ], 340 | "source": [ 341 | "#train and test loss and scores respectively\n", 342 | "train_loss_score=model.evaluate(x_train,y_train)\n", 343 | "test_loss_score=model.evaluate(x_test,y_test)\n", 344 | "print(train_loss_score)\n", 345 | "print(test_loss_score)" 346 | ] 347 | }, 348 | { 349 | "cell_type": "code", 350 | "execution_count": 14, 351 | "metadata": {}, 352 | "outputs": [ 353 | { 354 | "name": "stdout", 355 | "output_type": "stream", 356 | "text": [ 357 | "Model: \"sequential\"\n", 358 | "_________________________________________________________________\n", 359 | "Layer (type) Output Shape Param # \n", 360 | "=================================================================\n", 361 | "conv2d (Conv2D) (None, 40, 1, 64) 1664 \n", 362 | "_________________________________________________________________\n", 363 | "max_pooling2d (MaxPooling2D) (None, 20, 1, 64) 0 \n", 364 | "_________________________________________________________________\n", 365 | "conv2d_1 (Conv2D) (None, 20, 1, 128) 204928 \n", 366 | "_________________________________________________________________\n", 367 | "max_pooling2d_1 (MaxPooling2 (None, 10, 1, 128) 0 \n", 368 | "_________________________________________________________________\n", 369 | "dropout (Dropout) (None, 10, 1, 128) 0 \n", 370 | "_________________________________________________________________\n", 371 | "flatten (Flatten) (None, 1280) 0 \n", 372 | "_________________________________________________________________\n", 373 | "dense (Dense) (None, 256) 327936 \n", 374 | "_________________________________________________________________\n", 375 | "dropout_1 (Dropout) (None, 256) 0 \n", 376 | "_________________________________________________________________\n", 377 | "dense_1 (Dense) (None, 512) 131584 \n", 378 | "_________________________________________________________________\n", 379 | "dropout_2 (Dropout) (None, 512) 0 \n", 380 | "_________________________________________________________________\n", 381 | "dense_2 (Dense) (None, 10) 5130 \n", 382 | "=================================================================\n", 383 | "Total params: 671,242\n", 384 | "Trainable params: 671,242\n", 385 | "Non-trainable params: 0\n", 386 | "_________________________________________________________________\n" 387 | ] 388 | } 389 | ], 390 | "source": [ 391 | "model.summary()" 392 | ] 393 | }, 394 | { 395 | "cell_type": "code", 396 | "execution_count": 15, 397 | "metadata": {}, 398 | "outputs": [ 399 | { 400 | "name": "stdout", 401 | "output_type": "stream", 402 | "text": [ 403 | "WARNING:tensorflow:From /Users/vishrud/opt/anaconda3/envs/kerasCPU3/lib/python3.6/site-packages/tensorflow/python/ops/resource_variable_ops.py:1817: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.\n", 404 | "Instructions for updating:\n", 405 | "If using Keras pass *_constraint arguments to layers.\n", 406 | "INFO:tensorflow:Assets written to: urbanClassifier/assets\n" 407 | ] 408 | } 409 | ], 410 | "source": [ 411 | "model.save('urbanClassifier')" 412 | ] 413 | }, 414 | { 415 | "cell_type": "code", 416 | "execution_count": 16, 417 | "metadata": {}, 418 | "outputs": [], 419 | "source": [ 420 | "converter = tf.lite.TFLiteConverter.from_keras_model(model)\n", 421 | "tflite_model = converter.convert()\n", 422 | "\n", 423 | "# Save the TF Lite model.\n", 424 | "with tf.io.gfile.GFile('model.tflite', 'wb') as f:\n", 425 | " f.write(tflite_model)" 426 | ] 427 | }, 428 | { 429 | "cell_type": "code", 430 | "execution_count": 17, 431 | "metadata": {}, 432 | "outputs": [ 433 | { 434 | "data": { 435 | "text/plain": [ 436 | "(40, 1, 1)" 437 | ] 438 | }, 439 | "execution_count": 17, 440 | "metadata": {}, 441 | "output_type": "execute_result" 442 | } 443 | ], 444 | "source": [ 445 | "x_test[0].shape" 446 | ] 447 | }, 448 | { 449 | "cell_type": "code", 450 | "execution_count": 18, 451 | "metadata": {}, 452 | "outputs": [], 453 | "source": [ 454 | "testData = np.expand_dims(x_test[5],axis=0)" 455 | ] 456 | }, 457 | { 458 | "cell_type": "code", 459 | "execution_count": 19, 460 | "metadata": {}, 461 | "outputs": [ 462 | { 463 | "name": "stdout", 464 | "output_type": "stream", 465 | "text": [ 466 | "WARNING:tensorflow:From :1: Sequential.predict_classes (from tensorflow.python.keras.engine.sequential) is deprecated and will be removed after 2021-01-01.\n", 467 | "Instructions for updating:\n", 468 | "Please use instead:* `np.argmax(model.predict(x), axis=-1)`, if your model does multi-class classification (e.g. if it uses a `softmax` last-layer activation).* `(model.predict(x) > 0.5).astype(\"int32\")`, if your model does binary classification (e.g. if it uses a `sigmoid` last-layer activation).\n" 469 | ] 470 | }, 471 | { 472 | "name": "stderr", 473 | "output_type": "stream", 474 | "text": [ 475 | "WARNING:tensorflow:From :1: Sequential.predict_classes (from tensorflow.python.keras.engine.sequential) is deprecated and will be removed after 2021-01-01.\n", 476 | "Instructions for updating:\n", 477 | "Please use instead:* `np.argmax(model.predict(x), axis=-1)`, if your model does multi-class classification (e.g. if it uses a `softmax` last-layer activation).* `(model.predict(x) > 0.5).astype(\"int32\")`, if your model does binary classification (e.g. if it uses a `sigmoid` last-layer activation).\n" 478 | ] 479 | } 480 | ], 481 | "source": [ 482 | "prediction = model.predict_classes(testData)" 483 | ] 484 | }, 485 | { 486 | "cell_type": "code", 487 | "execution_count": 20, 488 | "metadata": {}, 489 | "outputs": [ 490 | { 491 | "data": { 492 | "text/plain": [ 493 | "array([3])" 494 | ] 495 | }, 496 | "execution_count": 20, 497 | "metadata": {}, 498 | "output_type": "execute_result" 499 | } 500 | ], 501 | "source": [ 502 | "prediction" 503 | ] 504 | }, 505 | { 506 | "cell_type": "code", 507 | "execution_count": 21, 508 | "metadata": {}, 509 | "outputs": [ 510 | { 511 | "data": { 512 | "text/plain": [ 513 | "array([0., 0., 0., 1., 0., 0., 0., 0., 0., 0.], dtype=float32)" 514 | ] 515 | }, 516 | "execution_count": 21, 517 | "metadata": {}, 518 | "output_type": "execute_result" 519 | } 520 | ], 521 | "source": [ 522 | "y_test[5]" 523 | ] 524 | }, 525 | { 526 | "cell_type": "code", 527 | "execution_count": 22, 528 | "metadata": {}, 529 | "outputs": [], 530 | "source": [ 531 | "# Load the TFLite model and allocate tensors.\n", 532 | "interpreter = tf.lite.Interpreter(model_path=\"model.tflite\")\n", 533 | "interpreter.allocate_tensors()\n", 534 | "\n", 535 | "# Get input and output tensors.\n", 536 | "input_details = interpreter.get_input_details()\n", 537 | "output_details = interpreter.get_output_details()" 538 | ] 539 | }, 540 | { 541 | "cell_type": "code", 542 | "execution_count": 23, 543 | "metadata": {}, 544 | "outputs": [ 545 | { 546 | "data": { 547 | "text/plain": [ 548 | "[{'name': 'conv2d_input',\n", 549 | " 'index': 0,\n", 550 | " 'shape': array([ 1, 40, 1, 1], dtype=int32),\n", 551 | " 'shape_signature': array([ 1, 40, 1, 1], dtype=int32),\n", 552 | " 'dtype': numpy.float32,\n", 553 | " 'quantization': (0.0, 0),\n", 554 | " 'quantization_parameters': {'scales': array([], dtype=float32),\n", 555 | " 'zero_points': array([], dtype=int32),\n", 556 | " 'quantized_dimension': 0},\n", 557 | " 'sparsity_parameters': {}}]" 558 | ] 559 | }, 560 | "execution_count": 23, 561 | "metadata": {}, 562 | "output_type": "execute_result" 563 | } 564 | ], 565 | "source": [ 566 | "input_details" 567 | ] 568 | }, 569 | { 570 | "cell_type": "code", 571 | "execution_count": 26, 572 | "metadata": {}, 573 | "outputs": [], 574 | "source": [ 575 | "testData = np.expand_dims(x_test[5],axis=0)\n", 576 | "atData = np.float32(testData)" 577 | ] 578 | }, 579 | { 580 | "cell_type": "code", 581 | "execution_count": 25, 582 | "metadata": {}, 583 | "outputs": [ 584 | { 585 | "name": "stdout", 586 | "output_type": "stream", 587 | "text": [ 588 | "float32\n", 589 | "(1, 40, 1, 1)\n", 590 | "[[7.1473175e-11 7.8754218e-11 7.2361290e-05 9.9991715e-01 4.1809503e-07\n", 591 | " 2.0054536e-08 2.4804867e-06 4.1848831e-09 7.6419092e-06 3.6501409e-09]]\n" 592 | ] 593 | } 594 | ], 595 | "source": [ 596 | "input_shape = input_details[0]['shape']\n", 597 | "input_data = np.array(np.random.random_sample(input_shape), dtype=np.float32)\n", 598 | "interpreter.set_tensor(input_details[0]['index'], atData)\n", 599 | "print(input_data.dtype)\n", 600 | "print(testData.shape)\n", 601 | "interpreter.invoke()\n", 602 | "\n", 603 | "# The function `get_tensor()` returns a copy of the tensor data.\n", 604 | "# Use `tensor()` in order to get a pointer to the tensor.\n", 605 | "output_data = interpreter.get_tensor(output_details[0]['index'])\n", 606 | "print(output_data)" 607 | ] 608 | }, 609 | { 610 | "cell_type": "code", 611 | "execution_count": null, 612 | "metadata": {}, 613 | "outputs": [], 614 | "source": [] 615 | } 616 | ], 617 | "metadata": { 618 | "accelerator": "GPU", 619 | "colab": { 620 | "collapsed_sections": [], 621 | "name": "Urban_cnn_model.ipynb", 622 | "provenance": [], 623 | "version": "0.3.2" 624 | }, 625 | "kernelspec": { 626 | "display_name": "kerasCPU3", 627 | "language": "python", 628 | "name": "kerascpu3" 629 | }, 630 | "language_info": { 631 | "codemirror_mode": { 632 | "name": "ipython", 633 | "version": 3 634 | }, 635 | "file_extension": ".py", 636 | "mimetype": "text/x-python", 637 | "name": "python", 638 | "nbconvert_exporter": "python", 639 | "pygments_lexer": "ipython3", 640 | "version": "3.6.10" 641 | } 642 | }, 643 | "nbformat": 4, 644 | "nbformat_minor": 1 645 | } 646 | -------------------------------------------------------------------------------- /Python_Notebook_Files/Urban_data_preprocess.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "code", 5 | "execution_count": 1, 6 | "metadata": { 7 | "colab": {}, 8 | "colab_type": "code", 9 | "id": "AUHre7lQ4eAF" 10 | }, 11 | "outputs": [], 12 | "source": [ 13 | "import pandas as pd\n", 14 | "import numpy as np\n", 15 | "import matplotlib.pyplot as plt\n", 16 | "import os\n", 17 | "from tqdm import tqdm" 18 | ] 19 | }, 20 | { 21 | "cell_type": "code", 22 | "execution_count": 2, 23 | "metadata": { 24 | "colab": {}, 25 | "colab_type": "code", 26 | "id": "VPLIlUJN4euz" 27 | }, 28 | "outputs": [], 29 | "source": [ 30 | "#forming a panda dataframe from the metadata file\n", 31 | "data=pd.read_csv(\"UrbanSound8K/metadata/UrbanSound_200.csv\")" 32 | ] 33 | }, 34 | { 35 | "cell_type": "code", 36 | "execution_count": 6, 37 | "metadata": { 38 | "colab": { 39 | "base_uri": "https://localhost:8080/", 40 | "height": 204 41 | }, 42 | "colab_type": "code", 43 | "id": "6Oa7rVd54g7A", 44 | "outputId": "5fcf3079-7b67-4f5a-bf04-64581579ec4b" 45 | }, 46 | "outputs": [ 47 | { 48 | "data": { 49 | "text/html": [ 50 | "
\n", 51 | "\n", 64 | "\n", 65 | " \n", 66 | " \n", 67 | " \n", 68 | " \n", 69 | " \n", 70 | " \n", 71 | " \n", 72 | " \n", 73 | " \n", 74 | " \n", 75 | " \n", 76 | " \n", 77 | " \n", 78 | " \n", 79 | " \n", 80 | " \n", 81 | " \n", 82 | " \n", 83 | " \n", 84 | " \n", 85 | " \n", 86 | " \n", 87 | " \n", 88 | " \n", 89 | " \n", 90 | " \n", 91 | " \n", 92 | " \n", 93 | " \n", 94 | " \n", 95 | " \n", 96 | " \n", 97 | " \n", 98 | " \n", 99 | " \n", 100 | " \n", 101 | " \n", 102 | " \n", 103 | " \n", 104 | " \n", 105 | " \n", 106 | " \n", 107 | " \n", 108 | " \n", 109 | " \n", 110 | " \n", 111 | " \n", 112 | " \n", 113 | " \n", 114 | " \n", 115 | " \n", 116 | " \n", 117 | " \n", 118 | " \n", 119 | " \n", 120 | " \n", 121 | " \n", 122 | " \n", 123 | " \n", 124 | " \n", 125 | " \n", 126 | " \n", 127 | " \n", 128 | " \n", 129 | " \n", 130 | " \n", 131 | " \n", 132 | " \n", 133 | " \n", 134 | " \n", 135 | "
slice_file_namefsIDstartendsaliencefoldclassIDclass
0100032-3-0-0.wav1000320.00.317551153dog_bark
1100263-2-0-117.wav10026358.562.500000152children_playing
2100263-2-0-121.wav10026360.564.500000152children_playing
3100263-2-0-126.wav10026363.067.000000152children_playing
4100263-2-0-137.wav10026368.572.500000152children_playing
\n", 136 | "
" 137 | ], 138 | "text/plain": [ 139 | " slice_file_name fsID start end salience fold classID \\\n", 140 | "0 100032-3-0-0.wav 100032 0.0 0.317551 1 5 3 \n", 141 | "1 100263-2-0-117.wav 100263 58.5 62.500000 1 5 2 \n", 142 | "2 100263-2-0-121.wav 100263 60.5 64.500000 1 5 2 \n", 143 | "3 100263-2-0-126.wav 100263 63.0 67.000000 1 5 2 \n", 144 | "4 100263-2-0-137.wav 100263 68.5 72.500000 1 5 2 \n", 145 | "\n", 146 | " class \n", 147 | "0 dog_bark \n", 148 | "1 children_playing \n", 149 | "2 children_playing \n", 150 | "3 children_playing \n", 151 | "4 children_playing " 152 | ] 153 | }, 154 | "execution_count": 6, 155 | "metadata": {}, 156 | "output_type": "execute_result" 157 | } 158 | ], 159 | "source": [ 160 | "#head of the dataframe\n", 161 | "data.head()" 162 | ] 163 | }, 164 | { 165 | "cell_type": "code", 166 | "execution_count": 12, 167 | "metadata": { 168 | "colab": { 169 | "base_uri": "https://localhost:8080/", 170 | "height": 204 171 | }, 172 | "colab_type": "code", 173 | "id": "WB0QggJl4imK", 174 | "outputId": "f71085a3-9ce5-491c-de8e-a48ca6d287bd" 175 | }, 176 | "outputs": [ 177 | { 178 | "data": { 179 | "text/plain": [ 180 | "10 50\n", 181 | "5 43\n", 182 | "1 31\n", 183 | "9 27\n", 184 | "2 23\n", 185 | "7 14\n", 186 | "3 8\n", 187 | "8 3\n", 188 | "6 3\n", 189 | "4 1\n", 190 | "Name: fold, dtype: int64" 191 | ] 192 | }, 193 | "execution_count": 12, 194 | "metadata": {}, 195 | "output_type": "execute_result" 196 | } 197 | ], 198 | "source": [ 199 | "#count of datapoints in each of the folders\n", 200 | "data[\"fold\"].value_counts()" 201 | ] 202 | }, 203 | { 204 | "cell_type": "code", 205 | "execution_count": 13, 206 | "metadata": { 207 | "colab": {}, 208 | "colab_type": "code", 209 | "id": "AXnks9vV4kct" 210 | }, 211 | "outputs": [], 212 | "source": [ 213 | "from librosa import display\n", 214 | "import librosa" 215 | ] 216 | }, 217 | { 218 | "cell_type": "code", 219 | "execution_count": 14, 220 | "metadata": { 221 | "colab": { 222 | "base_uri": "https://localhost:8080/", 223 | "height": 34 224 | }, 225 | "colab_type": "code", 226 | "id": "PrpVl_uJFtRz", 227 | "outputId": "72ac220b-c616-4bab-e040-b1f1d6be20a5" 228 | }, 229 | "outputs": [], 230 | "source": [ 231 | "#feature set\n", 232 | "#This file is of a dog bark\n", 233 | "y,sr=librosa.load(\"UrbanSound8K/audio/fold5/100032-3-0-0.wav\")\n", 234 | "mfccs = librosa.feature.mfcc(y, sr, n_mfcc=40)" 235 | ] 236 | }, 237 | { 238 | "cell_type": "code", 239 | "execution_count": 15, 240 | "metadata": {}, 241 | "outputs": [ 242 | { 243 | "name": "stdout", 244 | "output_type": "stream", 245 | "text": [ 246 | "(49440,)\n", 247 | "[3.0517578e-05 3.0517578e-05 0.0000000e+00 ... 1.6174316e-03 1.9226074e-03\n", 248 | " 2.5634766e-03]\n" 249 | ] 250 | } 251 | ], 252 | "source": [ 253 | "y,sr=librosa.load(\"UrbanSound8K/1995-1826-0003.wav\", sr=None)\n", 254 | "print(y.shape)\n", 255 | "print(y)\n", 256 | "#mfccs = librosa.feature.mfcc(y, sr, n_mfcc=40)" 257 | ] 258 | }, 259 | { 260 | "cell_type": "code", 261 | "execution_count": 16, 262 | "metadata": { 263 | "colab": { 264 | "base_uri": "https://localhost:8080/", 265 | "height": 293 266 | }, 267 | "colab_type": "code", 268 | "id": "MfAip77-F78F", 269 | "outputId": "1fd581e0-cad3-4444-ff81-91e749de3316" 270 | }, 271 | "outputs": [ 272 | { 273 | "data": { 274 | "image/png": "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\n", 275 | "text/plain": [ 276 | "
" 277 | ] 278 | }, 279 | "metadata": { 280 | "needs_background": "light" 281 | }, 282 | "output_type": "display_data" 283 | } 284 | ], 285 | "source": [ 286 | "#MFCC of dog bark\n", 287 | "import matplotlib.pyplot as plt\n", 288 | "plt.figure(figsize=(10,4))\n", 289 | "librosa.display.specshow(mfccs, x_axis='time')\n", 290 | "plt.colorbar()\n", 291 | "plt.title('MFCC')\n", 292 | "plt.tight_layout()" 293 | ] 294 | }, 295 | { 296 | "cell_type": "code", 297 | "execution_count": 17, 298 | "metadata": { 299 | "colab": {}, 300 | "colab_type": "code", 301 | "id": "b4xPM6yi4nxW" 302 | }, 303 | "outputs": [ 304 | { 305 | "name": "stderr", 306 | "output_type": "stream", 307 | "text": [ 308 | "100%|██████████| 203/203 [00:24<00:00, 8.16it/s]\n" 309 | ] 310 | } 311 | ], 312 | "source": [ 313 | "#preprocessing using only mfcc\n", 314 | "x_train=[]\n", 315 | "x_test=[]\n", 316 | "y_train=[]\n", 317 | "y_test=[]\n", 318 | "path=\"UrbanSound8K/audio/fold\"\n", 319 | "for i in tqdm(range(len(data))):\n", 320 | " fold_no=str(data.iloc[i][\"fold\"])\n", 321 | " file=data.iloc[i][\"slice_file_name\"]\n", 322 | " label=data.iloc[i][\"classID\"]\n", 323 | " filename=path+fold_no+\"/\"+file\n", 324 | " #print(filename)\n", 325 | " y,sr=librosa.load(filename)\n", 326 | " mfccs = np.mean(librosa.feature.mfcc(y, sr, n_mfcc=40).T,axis=0)\n", 327 | " #print(mfccs.shape,mfccs.max(),mfccs.min())\n", 328 | " if(fold_no!='10'):\n", 329 | " x_train.append(mfccs)\n", 330 | " y_train.append(label)\n", 331 | " else:\n", 332 | " x_test.append(mfccs)\n", 333 | " y_test.append(label)" 334 | ] 335 | }, 336 | { 337 | "cell_type": "code", 338 | "execution_count": 18, 339 | "metadata": { 340 | "colab": { 341 | "base_uri": "https://localhost:8080/", 342 | "height": 34 343 | }, 344 | "colab_type": "code", 345 | "id": "lGKy40lJ4pbJ", 346 | "outputId": "ba6c39b0-6e4e-4f92-9258-90dc34d9b8cc" 347 | }, 348 | "outputs": [ 349 | { 350 | "data": { 351 | "text/plain": [ 352 | "203" 353 | ] 354 | }, 355 | "execution_count": 18, 356 | "metadata": {}, 357 | "output_type": "execute_result" 358 | } 359 | ], 360 | "source": [ 361 | "len(x_train)+len(x_test)" 362 | ] 363 | }, 364 | { 365 | "cell_type": "code", 366 | "execution_count": 19, 367 | "metadata": { 368 | "colab": { 369 | "base_uri": "https://localhost:8080/", 370 | "height": 34 371 | }, 372 | "colab_type": "code", 373 | "id": "qm2H-y6X4rfa", 374 | "outputId": "adabc298-07c6-4b5d-c3e4-28e4c9a6f087" 375 | }, 376 | "outputs": [ 377 | { 378 | "data": { 379 | "text/plain": [ 380 | "203" 381 | ] 382 | }, 383 | "execution_count": 19, 384 | "metadata": {}, 385 | "output_type": "execute_result" 386 | } 387 | ], 388 | "source": [ 389 | "len(data)" 390 | ] 391 | }, 392 | { 393 | "cell_type": "code", 394 | "execution_count": 20, 395 | "metadata": { 396 | "colab": { 397 | "base_uri": "https://localhost:8080/", 398 | "height": 34 399 | }, 400 | "colab_type": "code", 401 | "id": "UZ67ZUVa4sy5", 402 | "outputId": "0e089e8f-16a2-4390-e8e0-e947401b6d91" 403 | }, 404 | "outputs": [ 405 | { 406 | "data": { 407 | "text/plain": [ 408 | "((153, 40), (50, 40), (153,), (50,))" 409 | ] 410 | }, 411 | "execution_count": 20, 412 | "metadata": {}, 413 | "output_type": "execute_result" 414 | } 415 | ], 416 | "source": [ 417 | "#converting the lists into numpy arrays\n", 418 | "x_train=np.array(x_train)\n", 419 | "x_test=np.array(x_test)\n", 420 | "y_train=np.array(y_train)\n", 421 | "y_test=np.array(y_test)\n", 422 | "x_train.shape,x_test.shape,y_train.shape,y_test.shape" 423 | ] 424 | }, 425 | { 426 | "cell_type": "code", 427 | "execution_count": 22, 428 | "metadata": { 429 | "colab": {}, 430 | "colab_type": "code", 431 | "id": "EJ2MbJzP41Xe" 432 | }, 433 | "outputs": [], 434 | "source": [ 435 | "#saving the data numpy arrays\n", 436 | "np.savetxt(\"train_data.csv\", x_train, delimiter=\",\")\n", 437 | "np.savetxt(\"test_data.csv\",x_test,delimiter=\",\")\n", 438 | "np.savetxt(\"train_labels.csv\",y_train,delimiter=\",\")\n", 439 | "np.savetxt(\"test_labels.csv\",y_test,delimiter=\",\")" 440 | ] 441 | }, 442 | { 443 | "cell_type": "code", 444 | "execution_count": 24, 445 | "metadata": {}, 446 | "outputs": [ 447 | { 448 | "data": { 449 | "text/plain": [ 450 | "(40,)" 451 | ] 452 | }, 453 | "execution_count": 24, 454 | "metadata": {}, 455 | "output_type": "execute_result" 456 | } 457 | ], 458 | "source": [ 459 | "x_test[0].shape" 460 | ] 461 | }, 462 | { 463 | "cell_type": "code", 464 | "execution_count": null, 465 | "metadata": {}, 466 | "outputs": [], 467 | "source": [] 468 | } 469 | ], 470 | "metadata": { 471 | "accelerator": "GPU", 472 | "colab": { 473 | "collapsed_sections": [], 474 | "name": "Urban_data_preprocess.ipynb", 475 | "provenance": [], 476 | "version": "0.3.2" 477 | }, 478 | "kernelspec": { 479 | "display_name": "kerasCPU3", 480 | "language": "python", 481 | "name": "kerascpu3" 482 | }, 483 | "language_info": { 484 | "codemirror_mode": { 485 | "name": "ipython", 486 | "version": 3 487 | }, 488 | "file_extension": ".py", 489 | "mimetype": "text/x-python", 490 | "name": "python", 491 | "nbconvert_exporter": "python", 492 | "pygments_lexer": "ipython3", 493 | "version": "3.6.10" 494 | } 495 | }, 496 | "nbformat": 4, 497 | "nbformat_minor": 1 498 | } 499 | -------------------------------------------------------------------------------- /Python_Notebook_Files/model.tflite: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/Python_Notebook_Files/model.tflite -------------------------------------------------------------------------------- /Python_Notebook_Files/test_labels.csv: -------------------------------------------------------------------------------- 1 | 1.000000000000000000e+00 2 | 1.000000000000000000e+00 3 | 1.000000000000000000e+00 4 | 1.000000000000000000e+00 5 | 1.000000000000000000e+00 6 | 3.000000000000000000e+00 7 | 3.000000000000000000e+00 8 | 3.000000000000000000e+00 9 | 3.000000000000000000e+00 10 | 2.000000000000000000e+00 11 | 2.000000000000000000e+00 12 | 2.000000000000000000e+00 13 | 2.000000000000000000e+00 14 | 2.000000000000000000e+00 15 | 2.000000000000000000e+00 16 | 2.000000000000000000e+00 17 | 2.000000000000000000e+00 18 | 3.000000000000000000e+00 19 | 3.000000000000000000e+00 20 | 5.000000000000000000e+00 21 | 5.000000000000000000e+00 22 | 5.000000000000000000e+00 23 | 5.000000000000000000e+00 24 | 5.000000000000000000e+00 25 | 5.000000000000000000e+00 26 | 5.000000000000000000e+00 27 | 5.000000000000000000e+00 28 | 5.000000000000000000e+00 29 | 5.000000000000000000e+00 30 | 5.000000000000000000e+00 31 | 5.000000000000000000e+00 32 | 5.000000000000000000e+00 33 | 5.000000000000000000e+00 34 | 5.000000000000000000e+00 35 | 5.000000000000000000e+00 36 | 5.000000000000000000e+00 37 | 5.000000000000000000e+00 38 | 5.000000000000000000e+00 39 | 5.000000000000000000e+00 40 | 5.000000000000000000e+00 41 | 5.000000000000000000e+00 42 | 5.000000000000000000e+00 43 | 5.000000000000000000e+00 44 | 5.000000000000000000e+00 45 | 5.000000000000000000e+00 46 | 5.000000000000000000e+00 47 | 5.000000000000000000e+00 48 | 5.000000000000000000e+00 49 | 5.000000000000000000e+00 50 | 5.000000000000000000e+00 51 | -------------------------------------------------------------------------------- /Python_Notebook_Files/train_labels.csv: -------------------------------------------------------------------------------- 1 | 3.000000000000000000e+00 2 | 2.000000000000000000e+00 3 | 2.000000000000000000e+00 4 | 2.000000000000000000e+00 5 | 2.000000000000000000e+00 6 | 2.000000000000000000e+00 7 | 2.000000000000000000e+00 8 | 2.000000000000000000e+00 9 | 2.000000000000000000e+00 10 | 3.000000000000000000e+00 11 | 3.000000000000000000e+00 12 | 3.000000000000000000e+00 13 | 3.000000000000000000e+00 14 | 0.000000000000000000e+00 15 | 0.000000000000000000e+00 16 | 0.000000000000000000e+00 17 | 0.000000000000000000e+00 18 | 0.000000000000000000e+00 19 | 0.000000000000000000e+00 20 | 0.000000000000000000e+00 21 | 0.000000000000000000e+00 22 | 0.000000000000000000e+00 23 | 0.000000000000000000e+00 24 | 0.000000000000000000e+00 25 | 0.000000000000000000e+00 26 | 0.000000000000000000e+00 27 | 0.000000000000000000e+00 28 | 0.000000000000000000e+00 29 | 0.000000000000000000e+00 30 | 0.000000000000000000e+00 31 | 0.000000000000000000e+00 32 | 0.000000000000000000e+00 33 | 0.000000000000000000e+00 34 | 0.000000000000000000e+00 35 | 0.000000000000000000e+00 36 | 0.000000000000000000e+00 37 | 0.000000000000000000e+00 38 | 0.000000000000000000e+00 39 | 0.000000000000000000e+00 40 | 0.000000000000000000e+00 41 | 0.000000000000000000e+00 42 | 0.000000000000000000e+00 43 | 0.000000000000000000e+00 44 | 0.000000000000000000e+00 45 | 3.000000000000000000e+00 46 | 3.000000000000000000e+00 47 | 3.000000000000000000e+00 48 | 3.000000000000000000e+00 49 | 3.000000000000000000e+00 50 | 3.000000000000000000e+00 51 | 0.000000000000000000e+00 52 | 0.000000000000000000e+00 53 | 0.000000000000000000e+00 54 | 0.000000000000000000e+00 55 | 0.000000000000000000e+00 56 | 0.000000000000000000e+00 57 | 0.000000000000000000e+00 58 | 0.000000000000000000e+00 59 | 0.000000000000000000e+00 60 | 0.000000000000000000e+00 61 | 0.000000000000000000e+00 62 | 0.000000000000000000e+00 63 | 0.000000000000000000e+00 64 | 0.000000000000000000e+00 65 | 0.000000000000000000e+00 66 | 0.000000000000000000e+00 67 | 0.000000000000000000e+00 68 | 0.000000000000000000e+00 69 | 0.000000000000000000e+00 70 | 0.000000000000000000e+00 71 | 0.000000000000000000e+00 72 | 0.000000000000000000e+00 73 | 0.000000000000000000e+00 74 | 0.000000000000000000e+00 75 | 0.000000000000000000e+00 76 | 0.000000000000000000e+00 77 | 0.000000000000000000e+00 78 | 9.000000000000000000e+00 79 | 9.000000000000000000e+00 80 | 9.000000000000000000e+00 81 | 9.000000000000000000e+00 82 | 9.000000000000000000e+00 83 | 9.000000000000000000e+00 84 | 3.000000000000000000e+00 85 | 3.000000000000000000e+00 86 | 3.000000000000000000e+00 87 | 3.000000000000000000e+00 88 | 6.000000000000000000e+00 89 | 3.000000000000000000e+00 90 | 3.000000000000000000e+00 91 | 3.000000000000000000e+00 92 | 3.000000000000000000e+00 93 | 3.000000000000000000e+00 94 | 3.000000000000000000e+00 95 | 3.000000000000000000e+00 96 | 8.000000000000000000e+00 97 | 8.000000000000000000e+00 98 | 8.000000000000000000e+00 99 | 8.000000000000000000e+00 100 | 8.000000000000000000e+00 101 | 8.000000000000000000e+00 102 | 8.000000000000000000e+00 103 | 8.000000000000000000e+00 104 | 5.000000000000000000e+00 105 | 5.000000000000000000e+00 106 | 8.000000000000000000e+00 107 | 8.000000000000000000e+00 108 | 8.000000000000000000e+00 109 | 8.000000000000000000e+00 110 | 8.000000000000000000e+00 111 | 8.000000000000000000e+00 112 | 8.000000000000000000e+00 113 | 8.000000000000000000e+00 114 | 8.000000000000000000e+00 115 | 8.000000000000000000e+00 116 | 8.000000000000000000e+00 117 | 8.000000000000000000e+00 118 | 8.000000000000000000e+00 119 | 8.000000000000000000e+00 120 | 8.000000000000000000e+00 121 | 8.000000000000000000e+00 122 | 7.000000000000000000e+00 123 | 7.000000000000000000e+00 124 | 7.000000000000000000e+00 125 | 7.000000000000000000e+00 126 | 7.000000000000000000e+00 127 | 7.000000000000000000e+00 128 | 7.000000000000000000e+00 129 | 7.000000000000000000e+00 130 | 7.000000000000000000e+00 131 | 7.000000000000000000e+00 132 | 7.000000000000000000e+00 133 | 7.000000000000000000e+00 134 | 7.000000000000000000e+00 135 | 7.000000000000000000e+00 136 | 7.000000000000000000e+00 137 | 7.000000000000000000e+00 138 | 7.000000000000000000e+00 139 | 7.000000000000000000e+00 140 | 7.000000000000000000e+00 141 | 7.000000000000000000e+00 142 | 7.000000000000000000e+00 143 | 7.000000000000000000e+00 144 | 3.000000000000000000e+00 145 | 3.000000000000000000e+00 146 | 3.000000000000000000e+00 147 | 4.000000000000000000e+00 148 | 4.000000000000000000e+00 149 | 4.000000000000000000e+00 150 | 4.000000000000000000e+00 151 | 4.000000000000000000e+00 152 | 4.000000000000000000e+00 153 | 4.000000000000000000e+00 154 | -------------------------------------------------------------------------------- /Readme.md: -------------------------------------------------------------------------------- 1 | Android app that helps in music classification based on Tensorflow models with the application of MFCC processing techniques. 2 | 3 | Refer this blog for the detailed overview: 4 | 5 | https://heartbeat.fritz.ai/noise-classification-in-android-mobile-app-with-tensorflow-lite-13dede0f5d44 6 | -------------------------------------------------------------------------------- /SampleAudio/001_children_playing.wav: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/SampleAudio/001_children_playing.wav -------------------------------------------------------------------------------- /SampleAudio/001_dog_barking.wav: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/SampleAudio/001_dog_barking.wav -------------------------------------------------------------------------------- /SampleAudio/001_street_music.wav: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/SampleAudio/001_street_music.wav -------------------------------------------------------------------------------- /SampleAudio/002_dog_barking.wav: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/SampleAudio/002_dog_barking.wav -------------------------------------------------------------------------------- /app/.DS_Store: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/app/.DS_Store -------------------------------------------------------------------------------- /app/.gitignore: -------------------------------------------------------------------------------- 1 | /build -------------------------------------------------------------------------------- /app/build.gradle: -------------------------------------------------------------------------------- 1 | apply plugin: 'com.android.application' 2 | apply plugin: 'kotlin-android' 3 | apply plugin: 'kotlin-android-extensions' 4 | 5 | android { 6 | compileSdkVersion 28 7 | buildToolsVersion "30.0.0" 8 | 9 | defaultConfig { 10 | applicationId "com.tensorflow.android.noiseclassifier" 11 | minSdkVersion 23 12 | targetSdkVersion 28 13 | versionCode 1 14 | versionName "1.0" 15 | 16 | testInstrumentationRunner "androidx.test.runner.AndroidJUnitRunner" 17 | } 18 | 19 | aaptOptions { 20 | noCompress "tflite" 21 | } 22 | 23 | buildTypes { 24 | release { 25 | minifyEnabled false 26 | proguardFiles getDefaultProguardFile('proguard-android-optimize.txt'), 'proguard-rules.pro' 27 | } 28 | } 29 | } 30 | 31 | dependencies { 32 | implementation fileTree(dir: "libs", include: ["*.jar"]) 33 | implementation "org.jetbrains.kotlin:kotlin-stdlib:$kotlin_version" 34 | implementation 'androidx.core:core-ktx:1.3.0' 35 | implementation 'androidx.appcompat:appcompat:1.1.0' 36 | implementation 'androidx.constraintlayout:constraintlayout:1.1.3' 37 | testImplementation 'junit:junit:4.12' 38 | androidTestImplementation 'androidx.test.ext:junit:1.1.1' 39 | androidTestImplementation 'androidx.test.espresso:espresso-core:3.2.0' 40 | implementation 'org.tensorflow:tensorflow-lite:0.0.0-nightly' 41 | implementation('org.tensorflow:tensorflow-lite-support:0.0.0-nightly') { changing = true } 42 | } -------------------------------------------------------------------------------- /app/proguard-rules.pro: -------------------------------------------------------------------------------- 1 | # Add project specific ProGuard rules here. 2 | # You can control the set of applied configuration files using the 3 | # proguardFiles setting in build.gradle. 4 | # 5 | # For more details, see 6 | # http://developer.android.com/guide/developing/tools/proguard.html 7 | 8 | # If your project uses WebView with JS, uncomment the following 9 | # and specify the fully qualified class name to the JavaScript interface 10 | # class: 11 | #-keepclassmembers class fqcn.of.javascript.interface.for.webview { 12 | # public *; 13 | #} 14 | 15 | # Uncomment this to preserve the line number information for 16 | # debugging stack traces. 17 | #-keepattributes SourceFile,LineNumberTable 18 | 19 | # If you keep the line number information, uncomment this to 20 | # hide the original source file name. 21 | #-renamesourcefileattribute SourceFile -------------------------------------------------------------------------------- /app/src/.DS_Store: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/app/src/.DS_Store -------------------------------------------------------------------------------- /app/src/androidTest/java/com/tensorflow/android/noiseclassifier/ExampleInstrumentedTest.kt: -------------------------------------------------------------------------------- 1 | package com.tensorflow.android.noiseclassifier 2 | 3 | import androidx.test.platform.app.InstrumentationRegistry 4 | import androidx.test.ext.junit.runners.AndroidJUnit4 5 | 6 | import org.junit.Test 7 | import org.junit.runner.RunWith 8 | 9 | import org.junit.Assert.* 10 | 11 | /** 12 | * Instrumented test, which will execute on an Android device. 13 | * 14 | * See [testing documentation](http://d.android.com/tools/testing). 15 | */ 16 | @RunWith(AndroidJUnit4::class) 17 | class ExampleInstrumentedTest { 18 | @Test 19 | fun useAppContext() { 20 | // Context of the app under test. 21 | val appContext = InstrumentationRegistry.getInstrumentation().targetContext 22 | assertEquals("com.tensorflow.android.noiseclassifier", appContext.packageName) 23 | } 24 | } -------------------------------------------------------------------------------- /app/src/main/.DS_Store: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/app/src/main/.DS_Store -------------------------------------------------------------------------------- /app/src/main/AndroidManifest.xml: -------------------------------------------------------------------------------- 1 | 2 | 4 | 5 | 6 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | -------------------------------------------------------------------------------- /app/src/main/assets/labels.txt: -------------------------------------------------------------------------------- 1 | air_conditioner, 2 | car_horn, 3 | children_playing, 4 | dog_bark, 5 | drilling, 6 | gun_shot, 7 | engine_idling, 8 | jackhammer, 9 | siren, 10 | street_music -------------------------------------------------------------------------------- /app/src/main/assets/model.tflite: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/app/src/main/assets/model.tflite -------------------------------------------------------------------------------- /app/src/main/java/.DS_Store: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/app/src/main/java/.DS_Store -------------------------------------------------------------------------------- /app/src/main/java/com/.DS_Store: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/app/src/main/java/com/.DS_Store -------------------------------------------------------------------------------- /app/src/main/java/com/tensorflow/.DS_Store: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/app/src/main/java/com/tensorflow/.DS_Store -------------------------------------------------------------------------------- /app/src/main/java/com/tensorflow/android/.DS_Store: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/VVasanth/Android_Tensorflow_AudioClassifier/3f8f58659d16e73405214b6bb400f6c156cab2a9/app/src/main/java/com/tensorflow/android/.DS_Store -------------------------------------------------------------------------------- /app/src/main/java/com/tensorflow/android/audio/features/FFT.java: -------------------------------------------------------------------------------- 1 | package com.tensorflow.android.audio.features; 2 | 3 | /** 4 | * Fast Fourier Transform. 5 | * 6 | * last updated on June 15, 2002
7 | * description: FFT class for real signals. Upon entry, N contains the 8 | * numbers of points in the DFT, real[] and imaginary[] contain the real and 9 | * imaginary parts of the input. Upon return, real[] and imaginary[] contain the 10 | * DFT output. All signals run from 0 to N - 1
11 | * input: speech signal
12 | * output: real and imaginary part of DFT output 13 | * 14 | * @author Danny Su 15 | * @author Hanns Holger Rutz 16 | */ 17 | public class FFT { 18 | double[] real; 19 | double[] imag; 20 | 21 | /** 22 | * Performs Fast Fourier Transformation in place. 23 | */ 24 | public void process(double[] signal) { 25 | final int numPoints = signal.length; 26 | // initialize real & imag array 27 | real = signal; 28 | imag = new double[numPoints]; 29 | 30 | // perform FFT using the real & imag array 31 | final double pi = Math.PI; 32 | final int numStages = (int) (Math.log(numPoints) / Math.log(2)); 33 | final int halfNumPoints = numPoints >> 1; 34 | int j = halfNumPoints; 35 | // FFT time domain decomposition carried out by "bit reversal sorting" 36 | // algorithm 37 | int k; 38 | for (int i = 1; i < numPoints - 2; i++) { 39 | if (i < j) { 40 | // swap 41 | double tempReal = real[j]; 42 | double tempImag = imag[j]; 43 | real[j] = real[i]; 44 | imag[j] = imag[i]; 45 | real[i] = tempReal; 46 | imag[i] = tempImag; 47 | } 48 | k = halfNumPoints; 49 | while (k <= j) { 50 | j -= k; 51 | k >>= 1; 52 | } 53 | j += k; 54 | } 55 | 56 | // loop for each stage 57 | for (int stage = 1; stage <= numStages; stage++) { 58 | int LE = 1; 59 | for (int i = 0; i < stage; i++) { 60 | LE <<= 1; 61 | } 62 | final int LE2 = LE >> 1; 63 | double UR = 1; 64 | double UI = 0; 65 | // calculate sine & cosine values 66 | final double SR = Math.cos(pi / LE2); 67 | final double SI = -Math.sin(pi / LE2); 68 | // loop for each sub DFT 69 | for (int subDFT = 1; subDFT <= LE2; subDFT++) { 70 | // loop for each butterfly 71 | for (int butterfly = subDFT - 1; butterfly <= numPoints - 1; butterfly += LE) { 72 | int ip = butterfly + LE2; 73 | // butterfly calculation 74 | double tempReal = (double) (real[ip] * UR - imag[ip] * UI); 75 | double tempImag = (double) (real[ip] * UI + imag[ip] * UR); 76 | real[ip] = real[butterfly] - tempReal; 77 | imag[ip] = imag[butterfly] - tempImag; 78 | real[butterfly] += tempReal; 79 | imag[butterfly] += tempImag; 80 | } 81 | 82 | double tempUR = UR; 83 | UR = tempUR * SR - UI * SI; 84 | UI = tempUR * SI + UI * SR; 85 | } 86 | } 87 | } 88 | } -------------------------------------------------------------------------------- /app/src/main/java/com/tensorflow/android/audio/features/MFCC.java: -------------------------------------------------------------------------------- 1 | package com.tensorflow.android.audio.features; 2 | 3 | /** 4 | * Wav file abstraction layer. 5 | * Source based on http://www.labbookpages.co.uk/audio/javaWavFiles.html 6 | * 7 | * 8 | */ 9 | 10 | public class MFCC { 11 | 12 | private static int n_mfcc = 40; 13 | private final static double fMin = 0.0; 14 | private final static int n_fft = 2048; 15 | private final static int hop_length = 512; 16 | private final static int n_mels = 128; 17 | 18 | private static double sampleRate = 44100.0; 19 | private final static double fMax = sampleRate/2.0; 20 | 21 | FFT fft = new FFT(); 22 | 23 | 24 | public void setSampleRate(int sampleRateVal){ 25 | sampleRate = sampleRateVal; 26 | } 27 | 28 | public void setN_mfcc(int n_mfccVal) { 29 | n_mfcc = n_mfccVal; 30 | } 31 | 32 | public float[] process(double[] doubleInputBuffer) { 33 | final double[][] mfccResult = dctMfcc(doubleInputBuffer); 34 | return finalshape(mfccResult); 35 | } 36 | 37 | //MFCC into 1d 38 | private float[] finalshape(double[][] mfccSpecTro){ 39 | float[] finalMfcc = new float[mfccSpecTro[0].length * mfccSpecTro.length]; 40 | int k = 0; 41 | for (int i = 0; i < mfccSpecTro[0].length; i++){ 42 | for (int j = 0; j < mfccSpecTro.length; j++){ 43 | finalMfcc[k] = (float) mfccSpecTro[j][i]; 44 | k = k+1; 45 | } 46 | } 47 | return finalMfcc; 48 | } 49 | 50 | //DCT to mfcc, librosa 51 | private double[][] dctMfcc(double[] y){ 52 | final double[][] specTroGram = powerToDb(melSpectrogram(y)); 53 | final double[][] dctBasis = dctFilter(n_mfcc, n_mels); 54 | double[][] mfccSpecTro = new double[n_mfcc][specTroGram[0].length]; 55 | for (int i = 0; i < n_mfcc; i++){ 56 | for (int j = 0; j < specTroGram[0].length; j++){ 57 | for (int k = 0; k < specTroGram.length; k++){ 58 | mfccSpecTro[i][j] += dctBasis[i][k]*specTroGram[k][j]; 59 | } 60 | } 61 | } 62 | return mfccSpecTro; 63 | } 64 | 65 | 66 | //mel spectrogram, librosa 67 | private double[][] melSpectrogram(double[] y){ 68 | double[][] melBasis = melFilter(); 69 | double[][] spectro = stftMagSpec(y); 70 | double[][] melS = new double[melBasis.length][spectro[0].length]; 71 | for (int i = 0; i < melBasis.length; i++){ 72 | for (int j = 0; j < spectro[0].length; j++){ 73 | for (int k = 0; k < melBasis[0].length; k++){ 74 | melS[i][j] += melBasis[i][k]*spectro[k][j]; 75 | } 76 | } 77 | } 78 | return melS; 79 | } 80 | 81 | 82 | //stft, librosa 83 | private double[][] stftMagSpec(double[] y){ 84 | //Short-time Fourier transform (STFT) 85 | final double[] fftwin = getWindow(); 86 | //pad y with reflect mode so it's centered. This reflect padding implementation is 87 | // not perfect but works for this demo. 88 | double[] ypad = new double[n_fft+y.length]; 89 | for (int i = 0; i < n_fft/2; i++){ 90 | ypad[(n_fft/2)-i-1] = y[i+1]; 91 | ypad[(n_fft/2)+y.length+i] = y[y.length-2-i]; 92 | } 93 | for (int j = 0; j < y.length; j++){ 94 | ypad[(n_fft/2)+j] = y[j]; 95 | } 96 | 97 | 98 | final double[][] frame = yFrame(ypad); 99 | double[][] fftmagSpec = new double[1+n_fft/2][frame[0].length]; 100 | double[] fftFrame = new double[n_fft]; 101 | for (int k = 0; k < frame[0].length; k++){ 102 | for (int l =0; l < n_fft; l++){ 103 | fftFrame[l] = fftwin[l]*frame[l][k]; 104 | } 105 | double[] magSpec = magSpectrogram(fftFrame); 106 | for (int i =0; i < 1+n_fft/2; i++){ 107 | fftmagSpec[i][k] = magSpec[i]; 108 | } 109 | } 110 | return fftmagSpec; 111 | } 112 | 113 | private double[] magSpectrogram(double[] frame){ 114 | double[] magSpec = new double[frame.length]; 115 | fft.process(frame); 116 | for (int m = 0; m < frame.length; m++) { 117 | magSpec[m] = fft.real[m] * fft.real[m] + fft.imag[m] * fft.imag[m]; 118 | } 119 | return magSpec; 120 | } 121 | 122 | 123 | //get hann window, librosa 124 | private double[] getWindow(){ 125 | //Return a Hann window for even n_fft. 126 | //The Hann window is a taper formed by using a raised cosine or sine-squared 127 | //with ends that touch zero. 128 | double[] win = new double[n_fft]; 129 | for (int i = 0; i < n_fft; i++){ 130 | win[i] = 0.5 - 0.5 * Math.cos(2.0*Math.PI*i/n_fft); 131 | } 132 | return win; 133 | } 134 | 135 | //frame, librosa 136 | private double[][] yFrame(double[] ypad){ 137 | final int n_frames = 1 + (ypad.length - n_fft) / hop_length; 138 | double[][] winFrames = new double[n_fft][n_frames]; 139 | for (int i = 0; i < n_fft; i++){ 140 | for (int j = 0; j < n_frames; j++){ 141 | winFrames[i][j] = ypad[j*hop_length+i]; 142 | } 143 | } 144 | return winFrames; 145 | } 146 | 147 | //power to db, librosa 148 | private double[][] powerToDb(double[][] melS){ 149 | //Convert a power spectrogram (amplitude squared) to decibel (dB) units 150 | // This computes the scaling ``10 * log10(S / ref)`` in a numerically 151 | // stable way. 152 | double[][] log_spec = new double[melS.length][melS[0].length]; 153 | double maxValue = -100; 154 | for (int i = 0; i < melS.length; i++){ 155 | for (int j = 0; j < melS[0].length; j++){ 156 | double magnitude = Math.abs(melS[i][j]); 157 | if (magnitude > 1e-10){ 158 | log_spec[i][j]=10.0*log10(magnitude); 159 | }else{ 160 | log_spec[i][j]=10.0*(-10); 161 | } 162 | if (log_spec[i][j] > maxValue){ 163 | maxValue = log_spec[i][j]; 164 | } 165 | } 166 | } 167 | 168 | //set top_db to 80.0 169 | for (int i = 0; i < melS.length; i++){ 170 | for (int j = 0; j < melS[0].length; j++){ 171 | if (log_spec[i][j] < maxValue - 80.0){ 172 | log_spec[i][j] = maxValue - 80.0; 173 | } 174 | } 175 | } 176 | //ref is disabled, maybe later. 177 | return log_spec; 178 | } 179 | 180 | //dct, librosa 181 | private double[][] dctFilter(int n_filters, int n_input){ 182 | //Discrete cosine transform (DCT type-III) basis. 183 | double[][] basis = new double[n_filters][n_input]; 184 | double[] samples = new double[n_input]; 185 | for (int i = 0; i < n_input; i++){ 186 | samples[i] = (1 + 2*i) * Math.PI/(2.0*(n_input)); 187 | } 188 | for (int j = 0; j < n_input; j++){ 189 | basis[0][j] = 1.0/Math.sqrt(n_input); 190 | } 191 | for (int i = 1; i < n_filters; i++){ 192 | for (int j = 0; j < n_input; j++){ 193 | basis[i][j] = Math.cos(i*samples[j]) * Math.sqrt(2.0/(n_input)); 194 | } 195 | } 196 | return basis; 197 | } 198 | 199 | 200 | //mel, librosa 201 | private double[][] melFilter(){ 202 | //Create a Filterbank matrix to combine FFT bins into Mel-frequency bins. 203 | // Center freqs of each FFT bin 204 | final double[] fftFreqs = fftFreq(); 205 | //'Center freqs' of mel bands - uniformly spaced between limits 206 | final double[] melF = melFreq(n_mels+2); 207 | 208 | double[] fdiff = new double[melF.length-1]; 209 | for (int i = 0; i < melF.length-1; i++){ 210 | fdiff[i] = melF[i+1]-melF[i]; 211 | } 212 | 213 | double[][] ramps = new double[melF.length][fftFreqs.length]; 214 | for (int i = 0; i < melF.length; i++){ 215 | for (int j = 0; j < fftFreqs.length; j++){ 216 | ramps[i][j] = melF[i]-fftFreqs[j]; 217 | } 218 | } 219 | 220 | double[][] weights = new double[n_mels][1+n_fft/2]; 221 | for (int i = 0; i < n_mels; i++){ 222 | for (int j = 0; j < fftFreqs.length; j++){ 223 | double lowerF = -ramps[i][j] / fdiff[i]; 224 | double upperF = ramps[i+2][j] / fdiff[i+1]; 225 | if (lowerF > upperF && upperF>0){ 226 | weights[i][j] = upperF; 227 | }else if (lowerF > upperF && upperF<0){ 228 | weights[i][j] = 0; 229 | }else if (lowerF < upperF && lowerF>0){ 230 | weights[i][j] =lowerF; 231 | }else if (lowerF < upperF && lowerF<0){ 232 | weights[i][j] = 0; 233 | }else {} 234 | } 235 | } 236 | 237 | double enorm[] = new double[n_mels]; 238 | for (int i = 0; i < n_mels; i++){ 239 | enorm[i] = 2.0 / (melF[i+2]-melF[i]); 240 | for (int j = 0; j < fftFreqs.length; j++){ 241 | weights[i][j] *= enorm[i]; 242 | } 243 | } 244 | return weights; 245 | 246 | //need to check if there's an empty channel somewhere 247 | } 248 | 249 | //fft frequencies, librosa 250 | private double[] fftFreq() { 251 | //Alternative implementation of np.fft.fftfreqs 252 | double[] freqs = new double[1+n_fft/2]; 253 | for (int i = 0; i < 1+n_fft/2; i++){ 254 | freqs[i] = 0 + (sampleRate/2)/(n_fft/2) * i; 255 | } 256 | return freqs; 257 | } 258 | 259 | //mel frequencies, librosa 260 | private double[] melFreq(int numMels) { 261 | //'Center freqs' of mel bands - uniformly spaced between limits 262 | double[] LowFFreq = new double[1]; 263 | double[] HighFFreq = new double[1]; 264 | LowFFreq[0] = fMin; 265 | HighFFreq[0] = fMax; 266 | final double[] melFLow = freqToMel(LowFFreq); 267 | final double[] melFHigh = freqToMel(HighFFreq); 268 | double[] mels = new double[numMels]; 269 | for (int i = 0; i < numMels; i++) { 270 | mels[i] = melFLow[0] + (melFHigh[0] - melFLow[0]) / (numMels-1) * i; 271 | } 272 | return melToFreq(mels); 273 | } 274 | 275 | 276 | //mel to hz, htk, librosa 277 | private double[] melToFreqS(double[] mels) { 278 | double[] freqs = new double[mels.length]; 279 | for (int i = 0; i < mels.length; i++) { 280 | freqs[i] = 700.0 * (Math.pow(10, mels[i]/2595.0) - 1.0); 281 | } 282 | return freqs; 283 | } 284 | 285 | 286 | // hz to mel, htk, librosa 287 | protected double[] freqToMelS(double[] freqs) { 288 | double[] mels = new double[freqs.length]; 289 | for (int i = 0; i < freqs.length; i++){ 290 | mels[i] = 2595.0 * log10(1.0 + freqs[i]/700.0); 291 | } 292 | return mels; 293 | } 294 | 295 | //mel to hz, Slaney, librosa 296 | private double[] melToFreq(double[] mels) { 297 | // Fill in the linear scale 298 | final double f_min = 0.0; 299 | final double f_sp = 200.0 / 3; 300 | double[] freqs = new double[mels.length]; 301 | 302 | // And now the nonlinear scale 303 | final double min_log_hz = 1000.0; // beginning of log region (Hz) 304 | final double min_log_mel = (min_log_hz - f_min) / f_sp; // same (Mels) 305 | final double logstep = Math.log(6.4) / 27.0; 306 | 307 | for (int i = 0; i < mels.length; i++) { 308 | if (mels[i] < min_log_mel){ 309 | freqs[i] = f_min + f_sp * mels[i]; 310 | }else{ 311 | freqs[i] = min_log_hz * Math.exp(logstep * (mels[i] - min_log_mel)); 312 | } 313 | } 314 | return freqs; 315 | } 316 | 317 | 318 | // hz to mel, Slaney, librosa 319 | protected double[] freqToMel(double[] freqs) { 320 | final double f_min = 0.0; 321 | final double f_sp = 200.0 / 3; 322 | double[] mels = new double[freqs.length]; 323 | 324 | // Fill in the log-scale part 325 | 326 | final double min_log_hz = 1000.0; // beginning of log region (Hz) 327 | final double min_log_mel = (min_log_hz - f_min) / f_sp ; // # same (Mels) 328 | final double logstep = Math.log(6.4) / 27.0; // step size for log region 329 | 330 | for (int i = 0; i < freqs.length; i++) { 331 | if (freqs[i] < min_log_hz){ 332 | mels[i] = (freqs[i] - f_min) / f_sp; 333 | }else{ 334 | mels[i] = min_log_mel + Math.log(freqs[i]/min_log_hz) / logstep; 335 | } 336 | } 337 | return mels; 338 | } 339 | 340 | // log10 341 | private double log10(double value) { 342 | return Math.log(value) / Math.log(10); 343 | } 344 | } 345 | -------------------------------------------------------------------------------- /app/src/main/java/com/tensorflow/android/audio/features/WavFile.java: -------------------------------------------------------------------------------- 1 | package com.tensorflow.android.audio.features; 2 | 3 | import java.io.ByteArrayOutputStream; 4 | import java.io.File; 5 | import java.io.FileInputStream; 6 | import java.io.FileOutputStream; 7 | import java.io.IOException; 8 | import java.math.RoundingMode; 9 | import java.text.DecimalFormat; 10 | 11 | /* 12 | Source based on https://github.com/Semantive/waveform-android/blob/master/library/src/main/java/com/semantive/waveformandroid/waveform/soundfile/WavFile.java 13 | */ 14 | 15 | public class WavFile { 16 | private enum IOState {READING, WRITING, CLOSED} 17 | 18 | private final static int BUFFER_SIZE = 4096; 19 | 20 | private final static int FMT_CHUNK_ID = 0x20746D66; 21 | private final static int DATA_CHUNK_ID = 0x61746164; 22 | private final static int RIFF_CHUNK_ID = 0x46464952; 23 | private final static int RIFF_TYPE_ID = 0x45564157; 24 | 25 | private File file; // File that will be read from or written to 26 | private IOState ioState; // Specifies the IO State of the Wav File (used for snaity checking) 27 | private int bytesPerSample; // Number of bytes required to store a single sample 28 | private long numFrames; // Number of frames within the data section 29 | private FileOutputStream oStream; // Output stream used for writting data 30 | private FileInputStream iStream; // Input stream used for reading data 31 | private float floatScale; // Scaling factor used for int <-> float conversion 32 | private float floatOffset; // Offset factor used for int <-> float conversion 33 | private boolean wordAlignAdjust; // Specify if an extra byte at the end of the data chunk is required for word alignment 34 | 35 | // Wav Header 36 | private int numChannels; // 2 bytes unsigned, 0x0001 (1) to 0xFFFF (65,535) 37 | private long sampleRate; // 4 bytes unsigned, 0x00000001 (1) to 0xFFFFFFFF (4,294,967,295) 38 | // Although a java int is 4 bytes, it is signed, so need to use a long 39 | private int blockAlign; // 2 bytes unsigned, 0x0001 (1) to 0xFFFF (65,535) 40 | private int validBits; // 2 bytes unsigned, 0x0002 (2) to 0xFFFF (65,535) 41 | 42 | // Buffering 43 | private byte[] buffer; // Local buffer used for IO 44 | private int bufferPointer; // Points to the current position in local buffer 45 | private int bytesRead; // Bytes read after last read into local buffer 46 | private long frameCounter; // Current number of frames read or written 47 | private long fileSize; 48 | 49 | // Cannot instantiate WavFile directly, must either use newWavFile() or openWavFile() 50 | private WavFile() { 51 | buffer = new byte[BUFFER_SIZE]; 52 | } 53 | 54 | public int getNumChannels() { 55 | return numChannels; 56 | } 57 | 58 | public long getNumFrames() { 59 | return numFrames; 60 | } 61 | 62 | public long getFramesRemaining() { 63 | return numFrames - frameCounter; 64 | } 65 | 66 | public long getSampleRate() { 67 | return sampleRate; 68 | } 69 | 70 | public int getValidBits() { 71 | return validBits; 72 | } 73 | 74 | public long getDuration() { 75 | return getNumFrames() / getSampleRate(); 76 | } 77 | 78 | public long getFileSize() { 79 | return fileSize; 80 | } 81 | 82 | public static WavFile openWavFile(File file) throws IOException, WavFileException { 83 | // Instantiate new Wavfile and store the file reference 84 | WavFile wavFile = new WavFile(); 85 | wavFile.file = file; 86 | 87 | // Create a new file input stream for reading file data 88 | wavFile.iStream = new FileInputStream(file); 89 | 90 | // Read the first 12 bytes of the file 91 | int bytesRead = wavFile.iStream.read(wavFile.buffer, 0, 12); 92 | if (bytesRead != 12) throw new WavFileException("Not enough wav file bytes for header"); 93 | 94 | // Extract parts from the header 95 | long riffChunkID = getLE(wavFile.buffer, 0, 4); 96 | long chunkSize = getLE(wavFile.buffer, 4, 4); 97 | long riffTypeID = getLE(wavFile.buffer, 8, 4); 98 | 99 | // Check the header bytes contains the correct signature 100 | if (riffChunkID != RIFF_CHUNK_ID) 101 | throw new WavFileException("Invalid Wav Header data, incorrect riff chunk ID"); 102 | if (riffTypeID != RIFF_TYPE_ID) throw new WavFileException("Invalid Wav Header data, incorrect riff type ID"); 103 | 104 | // Check that the file size matches the number of bytes listed in header 105 | if (file.length() != chunkSize + 8) { 106 | throw new WavFileException("Header chunk size (" + chunkSize + ") does not match file size (" + file.length() + ")"); 107 | } 108 | 109 | wavFile.fileSize = chunkSize; 110 | 111 | boolean foundFormat = false; 112 | boolean foundData = false; 113 | 114 | // Search for the Format and Data Chunks 115 | while (true) { 116 | // Read the first 8 bytes of the chunk (ID and chunk size) 117 | bytesRead = wavFile.iStream.read(wavFile.buffer, 0, 8); 118 | if (bytesRead == -1) throw new WavFileException("Reached end of file without finding format chunk"); 119 | if (bytesRead != 8) throw new WavFileException("Could not read chunk header"); 120 | 121 | // Extract the chunk ID and Size 122 | long chunkID = getLE(wavFile.buffer, 0, 4); 123 | chunkSize = getLE(wavFile.buffer, 4, 4); 124 | 125 | // Word align the chunk size 126 | // chunkSize specifies the number of bytes holding data. However, 127 | // the data should be word aligned (2 bytes) so we need to calculate 128 | // the actual number of bytes in the chunk 129 | long numChunkBytes = (chunkSize % 2 == 1) ? chunkSize + 1 : chunkSize; 130 | 131 | if (chunkID == FMT_CHUNK_ID) { 132 | // Flag that the format chunk has been found 133 | foundFormat = true; 134 | 135 | // Read in the header info 136 | bytesRead = wavFile.iStream.read(wavFile.buffer, 0, 16); 137 | 138 | // Check this is uncompressed data 139 | int compressionCode = (int) getLE(wavFile.buffer, 0, 2); 140 | if (compressionCode != 1) 141 | throw new WavFileException("Compression Code " + compressionCode + " not supported"); 142 | 143 | // Extract the format information 144 | wavFile.numChannels = (int) getLE(wavFile.buffer, 2, 2); 145 | wavFile.sampleRate = getLE(wavFile.buffer, 4, 4); 146 | wavFile.blockAlign = (int) getLE(wavFile.buffer, 12, 2); 147 | wavFile.validBits = (int) getLE(wavFile.buffer, 14, 2); 148 | 149 | if (wavFile.numChannels == 0) 150 | throw new WavFileException("Number of channels specified in header is equal to zero"); 151 | if (wavFile.blockAlign == 0) 152 | throw new WavFileException("Block Align specified in header is equal to zero"); 153 | if (wavFile.validBits < 2) throw new WavFileException("Valid Bits specified in header is less than 2"); 154 | if (wavFile.validBits > 64) 155 | throw new WavFileException("Valid Bits specified in header is greater than 64, this is greater than a long can hold"); 156 | 157 | // Calculate the number of bytes required to hold 1 sample 158 | wavFile.bytesPerSample = (wavFile.validBits + 7) / 8; 159 | if (wavFile.bytesPerSample * wavFile.numChannels != wavFile.blockAlign) 160 | throw new WavFileException("Block Align does not agree with bytes required for validBits and number of channels"); 161 | 162 | // Account for number of format bytes and then skip over 163 | // any extra format bytes 164 | numChunkBytes -= 16; 165 | if (numChunkBytes > 0) wavFile.iStream.skip(numChunkBytes); 166 | } else if (chunkID == DATA_CHUNK_ID) { 167 | // Check if we've found the format chunk, 168 | // If not, throw an exception as we need the format information 169 | // before we can read the data chunk 170 | if (foundFormat == false) throw new WavFileException("Data chunk found before Format chunk"); 171 | 172 | // Check that the chunkSize (wav data length) is a multiple of the 173 | // block align (bytes per frame) 174 | if (chunkSize % wavFile.blockAlign != 0) 175 | throw new WavFileException("Data Chunk size is not multiple of Block Align"); 176 | 177 | // Calculate the number of frames 178 | wavFile.numFrames = chunkSize / wavFile.blockAlign; 179 | 180 | // Flag that we've found the wave data chunk 181 | foundData = true; 182 | /* byte[] data = null; 183 | final ByteArrayOutputStream baout = new ByteArrayOutputStream(); 184 | int c; 185 | while ((c = wavFile.iStream.read(wavFile.buffer, 0, 16)) != -1) { 186 | baout.write(wavFile.buffer, 0, c); 187 | } 188 | 189 | baout.close(); 190 | data = baout.toByteArray(); 191 | 192 | double[] doubleInputBuffer = new double[data.length]; 193 | int j = 0; 194 | for (int i = 0; i < data.length; i++) { 195 | doubleInputBuffer[i] = data[i] / 32767.0; 196 | j = j + 1; 197 | } 198 | 199 | 200 | MFCC mfccConvert = new MFCC(); 201 | float[] mfccInput = mfccConvert.process(doubleInputBuffer); 202 | System.out.println(1); 203 | 204 | */ 205 | 206 | break; 207 | } else { 208 | // If an unknown chunk ID is found, just skip over the chunk data 209 | wavFile.iStream.skip(numChunkBytes); 210 | } 211 | } 212 | 213 | // Throw an exception if no data chunk has been found 214 | if (foundData == false) throw new WavFileException("Did not find a data chunk"); 215 | 216 | // Calculate the scaling factor for converting to a normalised double 217 | if (wavFile.validBits > 8) { 218 | // If more than 8 validBits, data is signed 219 | // Conversion required dividing by magnitude of max negative value 220 | wavFile.floatOffset = 0; 221 | wavFile.floatScale = 1 << (wavFile.validBits - 1); 222 | } else { 223 | // Else if 8 or less validBits, data is unsigned 224 | // Conversion required dividing by max positive value 225 | wavFile.floatOffset = -1; 226 | wavFile.floatScale = 0.5f * ((1 << wavFile.validBits) - 1); 227 | } 228 | 229 | wavFile.bufferPointer = 0; 230 | wavFile.bytesRead = 0; 231 | wavFile.frameCounter = 0; 232 | wavFile.ioState = IOState.READING; 233 | 234 | return wavFile; 235 | } 236 | 237 | private static long getLE(byte[] buffer, int pos, int numBytes) { 238 | numBytes--; 239 | pos += numBytes; 240 | 241 | long val = buffer[pos] & 0xFF; 242 | for (int b = 0; b < numBytes; b++) val = (val << 8) + (buffer[--pos] & 0xFF); 243 | 244 | return val; 245 | } 246 | 247 | private double readSample() throws IOException, WavFileException { 248 | long val = 0; 249 | 250 | for (int b = 0; b < bytesPerSample; b++) { 251 | if (bufferPointer == bytesRead) { 252 | int read = iStream.read(buffer, 0, BUFFER_SIZE); 253 | if (read == -1) throw new WavFileException("Not enough data available"); 254 | bytesRead = read; 255 | bufferPointer = 0; 256 | } 257 | 258 | int v = buffer[bufferPointer]; 259 | if (b < bytesPerSample - 1 || bytesPerSample == 1) v &= 0xFF; 260 | val += v << (b * 8); 261 | 262 | bufferPointer++; 263 | } 264 | 265 | return val/32767.0; 266 | } 267 | 268 | public int readFrames(float[] sampleBuffer, int numFramesToRead) throws IOException, WavFileException { 269 | return readFramesInternal(sampleBuffer, 0, numFramesToRead); 270 | } 271 | 272 | private int readFramesInternal(float[] sampleBuffer, int offset, int numFramesToRead) throws IOException, WavFileException { 273 | if (ioState != IOState.READING) throw new IOException("Cannot read from WavFile instance"); 274 | 275 | for (int f = 0; f < numFramesToRead; f++) { 276 | if (frameCounter == numFrames) return f; 277 | 278 | for (int c = 0; c < numChannels; c++) { 279 | sampleBuffer[offset] = floatOffset + (float) readSample() / floatScale; 280 | offset++; 281 | } 282 | 283 | frameCounter++; 284 | } 285 | 286 | return numFramesToRead; 287 | } 288 | 289 | public int readFrames(double[][] sampleBuffer, int numFramesToRead, int frameOffset) throws IOException, WavFileException { 290 | return readFramesInternal(sampleBuffer, frameOffset, numFramesToRead); 291 | } 292 | 293 | private int readFramesInternal(double[][] sampleBuffer, int frameOffset, int numFramesToRead) throws IOException, WavFileException { 294 | if (ioState != IOState.READING) throw new IOException("Cannot read from WavFile instance"); 295 | 296 | for (int f = 0; f < numFramesToRead; f++) { 297 | if (frameCounter == numFrames) return frameOffset; 298 | 299 | for (int c = 0; c < numChannels; c++) { 300 | sampleBuffer[c][frameOffset] = (double) readSample(); 301 | 302 | } 303 | frameCounter++; 304 | frameOffset++; 305 | 306 | } 307 | 308 | return frameOffset; 309 | } 310 | 311 | public void close() throws IOException { 312 | // Close the input stream and set to null 313 | if (iStream != null) { 314 | iStream.close(); 315 | iStream = null; 316 | } 317 | 318 | if (oStream != null) { 319 | // Write out anything still in the local buffer 320 | if (bufferPointer > 0) oStream.write(buffer, 0, bufferPointer); 321 | 322 | // If an extra byte is required for word alignment, add it to the end 323 | if (wordAlignAdjust) oStream.write(0); 324 | 325 | // Close the stream and set to null 326 | oStream.close(); 327 | oStream = null; 328 | } 329 | 330 | // Flag that the stream is closed 331 | ioState = IOState.CLOSED; 332 | } 333 | } -------------------------------------------------------------------------------- /app/src/main/java/com/tensorflow/android/audio/features/WavFileException.java: -------------------------------------------------------------------------------- 1 | package com.tensorflow.android.audio.features; 2 | 3 | public class WavFileException extends Exception { 4 | 5 | public WavFileException(final String message) { 6 | super(message); 7 | } 8 | } -------------------------------------------------------------------------------- /app/src/main/java/com/tensorflow/android/noiseclassifier/MainActivity.kt: -------------------------------------------------------------------------------- 1 | package com.tensorflow.android.noiseclassifier 2 | 3 | import android.os.Bundle 4 | import android.os.Environment 5 | import android.text.TextUtils 6 | import android.util.Log 7 | import android.view.View 8 | import android.widget.ArrayAdapter 9 | import android.widget.Spinner 10 | import android.widget.Toast 11 | import androidx.appcompat.app.AppCompatActivity 12 | import com.ml.quaterion.noiseClassification.Recognition 13 | import com.tensorflow.android.audio.features.MFCC 14 | import com.tensorflow.android.audio.features.WavFile 15 | import com.tensorflow.android.audio.features.WavFileException 16 | import com.tensorflow.android.noiseclassifier.R 17 | import kotlinx.android.synthetic.main.activity_main.* 18 | import org.tensorflow.lite.DataType 19 | import org.tensorflow.lite.Interpreter 20 | import org.tensorflow.lite.support.common.FileUtil 21 | import org.tensorflow.lite.support.common.TensorProcessor 22 | import org.tensorflow.lite.support.common.ops.NormalizeOp 23 | import org.tensorflow.lite.support.label.TensorLabel 24 | import org.tensorflow.lite.support.tensorbuffer.TensorBuffer 25 | import java.io.File 26 | import java.io.IOException 27 | import java.math.RoundingMode 28 | import java.nio.ByteBuffer 29 | import java.nio.MappedByteBuffer 30 | import java.text.DecimalFormat 31 | import java.util.* 32 | import kotlin.collections.ArrayList 33 | 34 | 35 | class MainActivity : AppCompatActivity() { 36 | 37 | 38 | override fun onCreate(savedInstanceState: Bundle?) { 39 | super.onCreate(savedInstanceState) 40 | setContentView(R.layout.activity_main) 41 | 42 | 43 | // val languages = resources.getStringArray(R.array.Languages) 44 | val externalStorage: File = Environment.getExternalStorageDirectory() 45 | 46 | val audioDirPath = externalStorage.absolutePath + "/audioData"; 47 | 48 | val fileNames: MutableList = ArrayList() 49 | 50 | 51 | File(audioDirPath).walk().forEach{ 52 | 53 | if(it.absolutePath.endsWith(".wav")){ 54 | fileNames.add(it.name) 55 | } 56 | 57 | } 58 | 59 | // access the spinner 60 | val spinner = findViewById(R.id.spinner) 61 | if (spinner != null) { 62 | val adapter = ArrayAdapter(this, 63 | android.R.layout.simple_spinner_item, fileNames) 64 | spinner.adapter = adapter 65 | 66 | } 67 | 68 | 69 | 70 | classify_button.setOnClickListener( View.OnClickListener { 71 | val selFilePath = spinner.selectedItem.toString() 72 | var audioFilePath = audioDirPath + '/' + selFilePath; 73 | if ( !TextUtils.isEmpty( selFilePath ) ){ 74 | 75 | 76 | val result = classifyNoise(audioFilePath) 77 | result_text.text = "Predicted Noise : $result" 78 | } 79 | else{ 80 | Toast.makeText( this@MainActivity, "Please enter a message.", Toast.LENGTH_LONG).show(); 81 | } 82 | }) 83 | 84 | } 85 | 86 | 87 | fun classifyNoise ( audioFilePath: String ): String? { 88 | 89 | val mNumFrames: Int 90 | val mSampleRate: Int 91 | val mChannels: Int 92 | var meanMFCCValues : FloatArray = FloatArray(1) 93 | 94 | var predictedResult: String? = "Unknown" 95 | 96 | var wavFile: WavFile? = null 97 | try { 98 | wavFile = WavFile.openWavFile(File(audioFilePath)) 99 | mNumFrames = wavFile.numFrames.toInt() 100 | mSampleRate = wavFile.sampleRate.toInt() 101 | mChannels = wavFile.numChannels 102 | val buffer = 103 | Array(mChannels) { DoubleArray(mNumFrames) } 104 | 105 | var frameOffset = 0 106 | val loopCounter: Int = mNumFrames * mChannels / 4096 + 1 107 | for (i in 0 until loopCounter) { 108 | frameOffset = wavFile.readFrames(buffer, mNumFrames, frameOffset) 109 | } 110 | 111 | //trimming the magnitude values to 5 decimal digits 112 | val df = DecimalFormat("#.#####") 113 | df.setRoundingMode(RoundingMode.CEILING) 114 | val meanBuffer = DoubleArray(mNumFrames) 115 | for (q in 0 until mNumFrames) { 116 | var frameVal = 0.0 117 | for (p in 0 until mChannels) { 118 | frameVal = frameVal + buffer[p][q] 119 | } 120 | meanBuffer[q] = df.format(frameVal / mChannels).toDouble() 121 | } 122 | 123 | 124 | //MFCC java library. 125 | val mfccConvert = MFCC() 126 | mfccConvert.setSampleRate(mSampleRate) 127 | val nMFCC = 40 128 | mfccConvert.setN_mfcc(nMFCC) 129 | val mfccInput = mfccConvert.process(meanBuffer) 130 | val nFFT = mfccInput.size / nMFCC 131 | val mfccValues = 132 | Array(nMFCC) { DoubleArray(nFFT) } 133 | 134 | //loop to convert the mfcc values into multi-dimensional array 135 | for (i in 0 until nFFT) { 136 | var indexCounter = i * nMFCC 137 | val rowIndexValue = i % nFFT 138 | for (j in 0 until nMFCC) { 139 | mfccValues[j][rowIndexValue] = mfccInput[indexCounter].toDouble() 140 | indexCounter++ 141 | } 142 | } 143 | 144 | //code to take the mean of mfcc values across the rows such that 145 | //[nMFCC x nFFT] matrix would be converted into 146 | //[nMFCC x 1] dimension - which would act as an input to tflite model 147 | meanMFCCValues = FloatArray(nMFCC) 148 | for (p in 0 until nMFCC) { 149 | var fftValAcrossRow = 0.0 150 | for (q in 0 until nFFT) { 151 | fftValAcrossRow = fftValAcrossRow + mfccValues[p][q] 152 | } 153 | val fftMeanValAcrossRow = fftValAcrossRow / nFFT 154 | meanMFCCValues[p] = fftMeanValAcrossRow.toFloat() 155 | } 156 | 157 | 158 | } catch (e: IOException) { 159 | e.printStackTrace() 160 | } catch (e: WavFileException) { 161 | e.printStackTrace() 162 | } 163 | 164 | predictedResult = loadModelAndMakePredictions(meanMFCCValues) 165 | 166 | return predictedResult 167 | 168 | } 169 | 170 | 171 | protected fun loadModelAndMakePredictions(meanMFCCValues : FloatArray) : String? { 172 | 173 | var predictedResult: String? = "unknown" 174 | 175 | //load the TFLite model in 'MappedByteBuffer' format using TF Interpreter 176 | val tfliteModel: MappedByteBuffer = 177 | FileUtil.loadMappedFile(this, getModelPath()) 178 | val tflite: Interpreter 179 | 180 | /** Options for configuring the Interpreter. */ 181 | val tfliteOptions = 182 | Interpreter.Options() 183 | tfliteOptions.setNumThreads(2) 184 | tflite = Interpreter(tfliteModel, tfliteOptions) 185 | 186 | //obtain the input and output tensor size required by the model 187 | //for urban sound classification, input tensor should be of 1x40x1x1 shape 188 | val imageTensorIndex = 0 189 | val imageShape = 190 | tflite.getInputTensor(imageTensorIndex).shape() 191 | val imageDataType: DataType = tflite.getInputTensor(imageTensorIndex).dataType() 192 | val probabilityTensorIndex = 0 193 | val probabilityShape = 194 | tflite.getOutputTensor(probabilityTensorIndex).shape() 195 | val probabilityDataType: DataType = 196 | tflite.getOutputTensor(probabilityTensorIndex).dataType() 197 | 198 | //need to transform the MFCC 1d float buffer into 1x40x1x1 dimension tensor using TensorBuffer 199 | val inBuffer: TensorBuffer = TensorBuffer.createDynamic(imageDataType) 200 | inBuffer.loadArray(meanMFCCValues, imageShape) 201 | val inpBuffer: ByteBuffer = inBuffer.getBuffer() 202 | val outputTensorBuffer: TensorBuffer = 203 | TensorBuffer.createFixedSize(probabilityShape, probabilityDataType) 204 | //run the predictions with input and output buffer tensors to get probability values across the labels 205 | tflite.run(inpBuffer, outputTensorBuffer.getBuffer()) 206 | 207 | 208 | //Code to transform the probability predictions into label values 209 | val ASSOCIATED_AXIS_LABELS = "labels.txt" 210 | var associatedAxisLabels: List? = null 211 | try { 212 | associatedAxisLabels = FileUtil.loadLabels(this, ASSOCIATED_AXIS_LABELS) 213 | } catch (e: IOException) { 214 | Log.e("tfliteSupport", "Error reading label file", e) 215 | } 216 | 217 | //Tensor processor for processing the probability values and to sort them based on the descending order of probabilities 218 | val probabilityProcessor: TensorProcessor = TensorProcessor.Builder() 219 | .add(NormalizeOp(0.0f, 255.0f)).build() 220 | if (null != associatedAxisLabels) { 221 | // Map of labels and their corresponding probability 222 | val labels = TensorLabel( 223 | associatedAxisLabels, 224 | probabilityProcessor.process(outputTensorBuffer) 225 | ) 226 | 227 | // Create a map to access the result based on label 228 | val floatMap: Map = 229 | labels.getMapWithFloatValue() 230 | 231 | //function to retrieve the top K probability values, in this case 'k' value is 1. 232 | //retrieved values are storied in 'Recognition' object with label details. 233 | val resultPrediction: List? = getTopKProbability(floatMap); 234 | 235 | //get the top 1 prediction from the retrieved list of top predictions 236 | predictedResult = getPredictedValue(resultPrediction) 237 | 238 | } 239 | return predictedResult 240 | 241 | } 242 | 243 | 244 | fun getPredictedValue(predictedList:List?): String?{ 245 | val top1PredictedValue : Recognition? = predictedList?.get(0) 246 | return top1PredictedValue?.getTitle() 247 | } 248 | 249 | fun getModelPath(): String { 250 | // you can download this file from 251 | // see build.gradle for where to obtain this file. It should be auto 252 | // downloaded into assets. 253 | return "model.tflite" 254 | } 255 | 256 | /** Gets the top-k results. */ 257 | protected fun getTopKProbability(labelProb: Map): List? { 258 | // Find the best classifications. 259 | val MAX_RESULTS: Int = 1 260 | val pq: PriorityQueue = PriorityQueue( 261 | MAX_RESULTS, 262 | Comparator { lhs, rhs -> // Intentionally reversed to put high confidence at the head of the queue. 263 | java.lang.Float.compare(rhs.getConfidence(), lhs.getConfidence()) 264 | }) 265 | for (entry in labelProb.entries) { 266 | pq.add(Recognition("" + entry.key, entry.key, entry.value)) 267 | } 268 | val recognitions: ArrayList = ArrayList() 269 | val recognitionsSize: Int = Math.min(pq.size, MAX_RESULTS) 270 | for (i in 0 until recognitionsSize) { 271 | recognitions.add(pq.poll()) 272 | } 273 | return recognitions 274 | } 275 | 276 | } 277 | -------------------------------------------------------------------------------- /app/src/main/java/com/tensorflow/android/noiseclassifier/Recognition.kt: -------------------------------------------------------------------------------- 1 | package com.ml.quaterion.noiseClassification 2 | 3 | 4 | class Recognition { 5 | 6 | 7 | private var id : String? = null 8 | private var title : String? = null 9 | private var confidence : Float 10 | 11 | constructor(id: String, title: String, confidence: Float){ 12 | 13 | this.id = id 14 | this.title = title 15 | this.confidence = confidence 16 | } 17 | 18 | fun getId(): String? { 19 | return id 20 | } 21 | 22 | fun getTitle(): String? { 23 | return title 24 | } 25 | 26 | fun getConfidence(): Float { 27 | return confidence 28 | } 29 | 30 | 31 | } -------------------------------------------------------------------------------- /app/src/main/res/drawable-v24/ic_launcher_foreground.xml: -------------------------------------------------------------------------------- 1 | 7 | 8 | 9 | 15 | 18 | 21 | 22 | 23 | 24 | 30 | -------------------------------------------------------------------------------- /app/src/main/res/drawable/ic_launcher_background.xml: -------------------------------------------------------------------------------- 1 | 2 | 7 | 10 | 15 | 20 | 25 | 30 | 35 | 40 | 45 | 50 | 55 | 60 | 65 | 70 | 75 | 80 | 85 | 90 | 95 | 100 | 105 | 110 | 115 | 120 | 125 | 130 | 135 | 140 | 145 | 150 | 155 | 160 | 165 | 170 | 171 | -------------------------------------------------------------------------------- /app/src/main/res/layout/activity_main.xml: -------------------------------------------------------------------------------- 1 | 2 | 8 | 9 | 17 |