├── .gitignore ├── About └── README.md ├── CODE_OF_CONDUCT.md ├── CONTRIBUTING.md ├── LICENSE ├── README.md ├── bin ├── stream.py ├── test.py ├── train.py └── utils.py ├── codecStatistic.py ├── codecTest.py ├── codecTrain.py ├── config ├── autoencoder │ ├── symAAD_vctk_48000_hop300.yaml │ ├── symAD_c16_vctk_48000_hop320.yaml │ ├── symAD_libritts_24000_hop300.yaml │ ├── symAD_vctk_48000_hop300.yaml │ └── symADuniv_vctk_48000_hop300.yaml ├── denoise │ └── symAD_vctk_48000_hop300.yaml ├── statistic │ ├── symAD_libritts_24000_hop300_clean.yaml │ ├── symAD_vctk_48000_hop300_clean.yaml │ └── symADuniv_vctk_48000_hop300_clean.yaml └── vocoder │ ├── AudioDec_v0_symAD_vctk_48000_hop300_clean.yaml │ ├── AudioDec_v1_symAD_libritts_24000_hop300_clean.yaml │ ├── AudioDec_v1_symAD_vctk_48000_hop300_clean.yaml │ ├── AudioDec_v2_symAD_vctk_48000_hop300_clean.yaml │ └── AudioDec_v3_symADuniv_vctk_48000_hop300_clean.yaml ├── dataloader ├── __init__.py ├── collater.py ├── dataset.py └── utils.py ├── demoFile.py ├── demoStream.py ├── exp ├── README.md ├── autoencoder │ ├── symAAD_vctk_48000_hop300 │ │ └── config.yml │ ├── symAD_c16_vctk_48000_hop320 │ │ └── config.yml │ ├── symAD_libritts_24000_hop300 │ │ └── config.yml │ ├── symAD_vctk_48000_hop300 │ │ └── config.yml │ └── symADuniv_vctk_48000_hop300 │ │ └── config.yml ├── denoise │ └── symAD_vctk_48000_hop300 │ │ └── config.yml └── vocoder │ ├── AudioDec_v0_symAD_vctk_48000_hop300_clean │ └── config.yml │ ├── AudioDec_v1_symAD_libritts_24000_hop300_clean │ └── config.yml │ ├── AudioDec_v1_symAD_vctk_48000_hop300_clean │ └── config.yml │ ├── AudioDec_v2_symAD_vctk_48000_hop300_clean │ └── config.yml │ └── AudioDec_v3_symADuniv_vctk_48000_hop300_clean │ └── config.yml ├── figs ├── architecture-1.png ├── latency.jpg └── mos.jpg ├── layers ├── activation_function.py ├── conv_layer.py └── vq_module.py ├── losses ├── __init__.py ├── adversarial_loss.py ├── feat_match_loss.py ├── mel_loss.py ├── stft_loss.py └── waveform_loss.py ├── models ├── autoencoder │ ├── AudioDec.py │ └── modules │ │ ├── decoder.py │ │ ├── encoder.py │ │ ├── projector.py │ │ ├── quantizer.py │ │ └── residual_unit.py ├── utils.py └── vocoder │ ├── HiFiGAN.py │ ├── UnivNet.py │ └── modules │ ├── discriminator.py │ ├── multi_fusion.py │ └── residual_block.py ├── parse_options.sh ├── requirements.txt ├── slurmlogs └── README.md ├── stats ├── symAD_libritts_24000_hop300_clean.npy ├── symAD_vctk_48000_hop300_clean.npy └── symADuniv_vctk_48000_hop300_clean.npy ├── submit_autoencoder.sh ├── submit_codecUniv_vctk.sh ├── submit_codec_libritts.sh ├── submit_codec_vctk.sh ├── submit_denoise.sh ├── submit_statistic.sh ├── submit_vocoder.sh ├── trainer ├── autoencoder.py ├── denoise.py ├── trainerGAN.py └── vocoder.py └── utils └── audiodec.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/.gitignore -------------------------------------------------------------------------------- /About/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/About/README.md -------------------------------------------------------------------------------- /CODE_OF_CONDUCT.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/CODE_OF_CONDUCT.md -------------------------------------------------------------------------------- /CONTRIBUTING.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/CONTRIBUTING.md -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/README.md -------------------------------------------------------------------------------- /bin/stream.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/bin/stream.py -------------------------------------------------------------------------------- /bin/test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/bin/test.py -------------------------------------------------------------------------------- /bin/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/bin/train.py -------------------------------------------------------------------------------- /bin/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/bin/utils.py -------------------------------------------------------------------------------- /codecStatistic.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/codecStatistic.py -------------------------------------------------------------------------------- /codecTest.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/codecTest.py -------------------------------------------------------------------------------- /codecTrain.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/codecTrain.py -------------------------------------------------------------------------------- /config/autoencoder/symAAD_vctk_48000_hop300.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/autoencoder/symAAD_vctk_48000_hop300.yaml -------------------------------------------------------------------------------- /config/autoencoder/symAD_c16_vctk_48000_hop320.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/autoencoder/symAD_c16_vctk_48000_hop320.yaml -------------------------------------------------------------------------------- /config/autoencoder/symAD_libritts_24000_hop300.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/autoencoder/symAD_libritts_24000_hop300.yaml -------------------------------------------------------------------------------- /config/autoencoder/symAD_vctk_48000_hop300.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/autoencoder/symAD_vctk_48000_hop300.yaml -------------------------------------------------------------------------------- /config/autoencoder/symADuniv_vctk_48000_hop300.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/autoencoder/symADuniv_vctk_48000_hop300.yaml -------------------------------------------------------------------------------- /config/denoise/symAD_vctk_48000_hop300.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/denoise/symAD_vctk_48000_hop300.yaml -------------------------------------------------------------------------------- /config/statistic/symAD_libritts_24000_hop300_clean.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/statistic/symAD_libritts_24000_hop300_clean.yaml -------------------------------------------------------------------------------- /config/statistic/symAD_vctk_48000_hop300_clean.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/statistic/symAD_vctk_48000_hop300_clean.yaml -------------------------------------------------------------------------------- /config/statistic/symADuniv_vctk_48000_hop300_clean.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/statistic/symADuniv_vctk_48000_hop300_clean.yaml -------------------------------------------------------------------------------- /config/vocoder/AudioDec_v0_symAD_vctk_48000_hop300_clean.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/vocoder/AudioDec_v0_symAD_vctk_48000_hop300_clean.yaml -------------------------------------------------------------------------------- /config/vocoder/AudioDec_v1_symAD_libritts_24000_hop300_clean.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/vocoder/AudioDec_v1_symAD_libritts_24000_hop300_clean.yaml -------------------------------------------------------------------------------- /config/vocoder/AudioDec_v1_symAD_vctk_48000_hop300_clean.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/vocoder/AudioDec_v1_symAD_vctk_48000_hop300_clean.yaml -------------------------------------------------------------------------------- /config/vocoder/AudioDec_v2_symAD_vctk_48000_hop300_clean.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/vocoder/AudioDec_v2_symAD_vctk_48000_hop300_clean.yaml -------------------------------------------------------------------------------- /config/vocoder/AudioDec_v3_symADuniv_vctk_48000_hop300_clean.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/config/vocoder/AudioDec_v3_symADuniv_vctk_48000_hop300_clean.yaml -------------------------------------------------------------------------------- /dataloader/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/dataloader/__init__.py -------------------------------------------------------------------------------- /dataloader/collater.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/dataloader/collater.py -------------------------------------------------------------------------------- /dataloader/dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/dataloader/dataset.py -------------------------------------------------------------------------------- /dataloader/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/dataloader/utils.py -------------------------------------------------------------------------------- /demoFile.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/demoFile.py -------------------------------------------------------------------------------- /demoStream.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/demoStream.py -------------------------------------------------------------------------------- /exp/README.md: -------------------------------------------------------------------------------- 1 | # Folder for saving models -------------------------------------------------------------------------------- /exp/autoencoder/symAAD_vctk_48000_hop300/config.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/exp/autoencoder/symAAD_vctk_48000_hop300/config.yml -------------------------------------------------------------------------------- /exp/autoencoder/symAD_c16_vctk_48000_hop320/config.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/exp/autoencoder/symAD_c16_vctk_48000_hop320/config.yml -------------------------------------------------------------------------------- /exp/autoencoder/symAD_libritts_24000_hop300/config.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/exp/autoencoder/symAD_libritts_24000_hop300/config.yml -------------------------------------------------------------------------------- /exp/autoencoder/symAD_vctk_48000_hop300/config.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/exp/autoencoder/symAD_vctk_48000_hop300/config.yml -------------------------------------------------------------------------------- /exp/autoencoder/symADuniv_vctk_48000_hop300/config.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/exp/autoencoder/symADuniv_vctk_48000_hop300/config.yml -------------------------------------------------------------------------------- /exp/denoise/symAD_vctk_48000_hop300/config.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/exp/denoise/symAD_vctk_48000_hop300/config.yml -------------------------------------------------------------------------------- /exp/vocoder/AudioDec_v0_symAD_vctk_48000_hop300_clean/config.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/exp/vocoder/AudioDec_v0_symAD_vctk_48000_hop300_clean/config.yml -------------------------------------------------------------------------------- /exp/vocoder/AudioDec_v1_symAD_libritts_24000_hop300_clean/config.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/exp/vocoder/AudioDec_v1_symAD_libritts_24000_hop300_clean/config.yml -------------------------------------------------------------------------------- /exp/vocoder/AudioDec_v1_symAD_vctk_48000_hop300_clean/config.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/exp/vocoder/AudioDec_v1_symAD_vctk_48000_hop300_clean/config.yml -------------------------------------------------------------------------------- /exp/vocoder/AudioDec_v2_symAD_vctk_48000_hop300_clean/config.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/exp/vocoder/AudioDec_v2_symAD_vctk_48000_hop300_clean/config.yml -------------------------------------------------------------------------------- /exp/vocoder/AudioDec_v3_symADuniv_vctk_48000_hop300_clean/config.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/exp/vocoder/AudioDec_v3_symADuniv_vctk_48000_hop300_clean/config.yml -------------------------------------------------------------------------------- /figs/architecture-1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/figs/architecture-1.png -------------------------------------------------------------------------------- /figs/latency.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/figs/latency.jpg -------------------------------------------------------------------------------- /figs/mos.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/figs/mos.jpg -------------------------------------------------------------------------------- /layers/activation_function.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/layers/activation_function.py -------------------------------------------------------------------------------- /layers/conv_layer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/layers/conv_layer.py -------------------------------------------------------------------------------- /layers/vq_module.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/layers/vq_module.py -------------------------------------------------------------------------------- /losses/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/losses/__init__.py -------------------------------------------------------------------------------- /losses/adversarial_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/losses/adversarial_loss.py -------------------------------------------------------------------------------- /losses/feat_match_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/losses/feat_match_loss.py -------------------------------------------------------------------------------- /losses/mel_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/losses/mel_loss.py -------------------------------------------------------------------------------- /losses/stft_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/losses/stft_loss.py -------------------------------------------------------------------------------- /losses/waveform_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/losses/waveform_loss.py -------------------------------------------------------------------------------- /models/autoencoder/AudioDec.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/autoencoder/AudioDec.py -------------------------------------------------------------------------------- /models/autoencoder/modules/decoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/autoencoder/modules/decoder.py -------------------------------------------------------------------------------- /models/autoencoder/modules/encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/autoencoder/modules/encoder.py -------------------------------------------------------------------------------- /models/autoencoder/modules/projector.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/autoencoder/modules/projector.py -------------------------------------------------------------------------------- /models/autoencoder/modules/quantizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/autoencoder/modules/quantizer.py -------------------------------------------------------------------------------- /models/autoencoder/modules/residual_unit.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/autoencoder/modules/residual_unit.py -------------------------------------------------------------------------------- /models/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/utils.py -------------------------------------------------------------------------------- /models/vocoder/HiFiGAN.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/vocoder/HiFiGAN.py -------------------------------------------------------------------------------- /models/vocoder/UnivNet.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/vocoder/UnivNet.py -------------------------------------------------------------------------------- /models/vocoder/modules/discriminator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/vocoder/modules/discriminator.py -------------------------------------------------------------------------------- /models/vocoder/modules/multi_fusion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/vocoder/modules/multi_fusion.py -------------------------------------------------------------------------------- /models/vocoder/modules/residual_block.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/models/vocoder/modules/residual_block.py -------------------------------------------------------------------------------- /parse_options.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/parse_options.sh -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/requirements.txt -------------------------------------------------------------------------------- /slurmlogs/README.md: -------------------------------------------------------------------------------- 1 | # Folder for saving slurm logs -------------------------------------------------------------------------------- /stats/symAD_libritts_24000_hop300_clean.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/stats/symAD_libritts_24000_hop300_clean.npy -------------------------------------------------------------------------------- /stats/symAD_vctk_48000_hop300_clean.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/stats/symAD_vctk_48000_hop300_clean.npy -------------------------------------------------------------------------------- /stats/symADuniv_vctk_48000_hop300_clean.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/stats/symADuniv_vctk_48000_hop300_clean.npy -------------------------------------------------------------------------------- /submit_autoencoder.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/submit_autoencoder.sh -------------------------------------------------------------------------------- /submit_codecUniv_vctk.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/submit_codecUniv_vctk.sh -------------------------------------------------------------------------------- /submit_codec_libritts.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/submit_codec_libritts.sh -------------------------------------------------------------------------------- /submit_codec_vctk.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/submit_codec_vctk.sh -------------------------------------------------------------------------------- /submit_denoise.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/submit_denoise.sh -------------------------------------------------------------------------------- /submit_statistic.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/submit_statistic.sh -------------------------------------------------------------------------------- /submit_vocoder.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/submit_vocoder.sh -------------------------------------------------------------------------------- /trainer/autoencoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/trainer/autoencoder.py -------------------------------------------------------------------------------- /trainer/denoise.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/trainer/denoise.py -------------------------------------------------------------------------------- /trainer/trainerGAN.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/trainer/trainerGAN.py -------------------------------------------------------------------------------- /trainer/vocoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/trainer/vocoder.py -------------------------------------------------------------------------------- /utils/audiodec.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/facebookresearch/AudioDec/HEAD/utils/audiodec.py --------------------------------------------------------------------------------