├── README.md
├── data.json
├── input.txt
└── tts.py


/README.md:
--------------------------------------------------------------------------------
 1 | # TTS-Grabber
 2 | Quick thing i made about a year ago to download any text with any tts voice, 829 voices to choose from currently.
 3 | 
 4 | The program will split the input into multiple files every 1500 words or so to not hit any cutoff limits from TTS providers.
 5 | 
 6 | ## Usage:  
 7 | Edit `input.txt` to change the text to synthesize.
 8 | 
 9 | You can run just `tts.py` without any parameters to open the voice selector with default settings.  
10 | 
11 | #### Parameters
12 | ```
13 | PARAMETER           TYPE  DESCRIPTION
14 | -h, -help           ---   Shows the help info.
15 | -v, -voice          Int   Sets the voice id to use.  
16 | -s, -speed          Int   Sets the TTS voice speed (in percent).  
17 | -vol, -volume       Int   Changes the TTS volume (in decibels).  
18 | -pp, -period-pause  Flt   Sets how long the TTS should pause for at periods (in seconds).  
19 | -cp, -comma-pause   Flt   Sets how long the TTS should pause for at commas (in seconds).  
20 | -lp, -line-pause    Flt   Sets how long the TTS should pause for at newlines (in seconds).
21 | ```
22 | 
23 | Example with parameters:  
24 | `tts.py -v 777 -s 100 -vol 0 -pp 1 -cp 0.5 -lp 2`
25 | 
26 | ###### absolutely no api abuse here
27 | 


--------------------------------------------------------------------------------
/input.txt:
--------------------------------------------------------------------------------
1 | Here are <say-as interpret-as="characters">SSML</say-as> samples.
2 | I can pause <break time="3s"/>.
3 | I can play a sound
4 | <audio src="https://download.samplelib.com/mp3/sample-3s.mp3">didn't get your MP3 audio file</audio>.
5 | I can speak in cardinals. Your number is <say-as interpret-as="cardinal">10</say-as>.
6 | Or I can speak in ordinals. You are <say-as interpret-as="ordinal">10</say-as> in line.
7 | Or I can even speak in digits. The digits for ten are <say-as interpret-as="characters">10</say-as>.
8 | I can also substitute phrases, like the <sub alias="World Wide Web Consortium">W3C</sub>.
9 | 


--------------------------------------------------------------------------------
/tts.py:
--------------------------------------------------------------------------------
  1 | import functools
  2 | import requests
  3 | import textwrap
  4 | import os.path
  5 | import time
  6 | import json
  7 | import sys
  8 | 
  9 | data = []
 10 | lastUsed = {}
 11 | 
 12 | def compare(dic1, dic2):
 13 |     s1 = dic1["language"] + dic1["gender"] + dic1["name"]
 14 |     s2 = dic2["language"] + dic2["gender"] + dic2["name"]
 15 | 
 16 |     if s1 > s2:
 17 |         return 1
 18 |     elif s2 > s1:
 19 |         return -1
 20 |     else:
 21 |         return 0
 22 | 
 23 | if len(sys.argv) == 2 and sys.argv[1].lower().startswith("-h"):
 24 |     print(
 25 | """TTS-Grabber (https://github.com/BleachDev/TTS-Grabber)
 26 | 
 27 | PARAMETER           TYPE  DESCRIPTION
 28 | -h, -help           ---   Shows the help info.
 29 | -v, -voice          Int   Sets the voice id to use.
 30 | -s, -speed          Int   Sets the TTS voice speed (in percent).
 31 | -vol, -volume       Int   Changes the TTS volume (in decibels).
 32 | -pp, -period-pause  Flt   Sets how long the TTS should pause for at periods (in seconds).
 33 | -cp, -comma-pause   Flt   Sets how long the TTS should pause for at commas (in seconds).
 34 | -lp, -line-pause    Flt   Sets how long the TTS should pause for at newlines (in seconds).
 35 | 
 36 | To see a list of the voices available, run the script without the -v parameter.""")
 37 |     sys.exit()
 38 |     
 39 | 
 40 | arg_voice = -1
 41 | arg_speed = 100
 42 | arg_volume = 0
 43 | arg_period_pause = -1
 44 | arg_comma_pause = -1
 45 | arg_line_pause = -1
 46 | 
 47 | _loop = 1
 48 | while _loop <= len(sys.argv) - 2:
 49 |     try:
 50 |         arg = sys.argv[_loop].lower()
 51 |         if arg == "-v" or arg == "-voice":
 52 |             arg_voice = int(sys.argv[_loop + 1])
 53 |         elif arg == "-s" or arg == "-speed":
 54 |             arg_speed = int(sys.argv[_loop + 1])
 55 |         elif arg == "-vol" or arg == "-volume":
 56 |             arg_volume = float(sys.argv[_loop + 1])
 57 |         elif arg == "-pp" or arg == "-period-pause":
 58 |             arg_period_pause = float(sys.argv[_loop + 1])
 59 |         elif arg == "-cp" or arg == "-comma-pause":
 60 |             arg_comma_pause = float(sys.argv[_loop + 1])
 61 |         elif arg == "-lp" or arg == "-line-pause":
 62 |             arg_line_pause = float(sys.argv[_loop + 1])
 63 |     except:
 64 |         print("error > " + sys.argv[_loop] + " / " + sys.argv[_loop + 1])
 65 |         pass
 66 |     _loop += 2
 67 | 
 68 | if os.path.isfile("lastused.json"):
 69 |     with open('lastused.json') as f:
 70 |         lastUsed = json.load(f)
 71 | 
 72 | with open('data.json') as f:
 73 |     js_data = json.load(f)
 74 | 
 75 | for entry in js_data:
 76 |     data.append(entry)
 77 | 
 78 | data.sort(key=functools.cmp_to_key(compare))
 79 | 
 80 | if arg_voice == -1:
 81 |     print("ID   LANGUAGE                        GENDER  NAME                TYPE")
 82 |     for i in range(len(data)):
 83 |         print(
 84 |             (" " * (3 - len(str(i + 1)))) + str(i + 1)
 85 |             + ": " + data[i]["language"] + (" " * (32 - len(data[i]["language"])))
 86 |             + data[i]["gender"] + (" " * (8 - len(data[i]["gender"])))
 87 |             + data[i]["name"] + (" " * (20 - len(data[i]["name"])))
 88 |             + data[i]["voiceType"][0])
 89 | 
 90 |     if lastUsed != {}:
 91 |         print("...........................................................")
 92 |         print("  0: " + lastUsed["language"] + (" " * (32 - len(lastUsed["language"])))
 93 |             + lastUsed["gender"] + (" " * (8 - len(lastUsed["gender"])))
 94 |             + lastUsed["name"] + (" " * (20 - len(lastUsed["name"])))
 95 |             + lastUsed["voiceType"][0])
 96 | 
 97 |     print("...........................................................")
 98 |     arg_voice = int(input("Choose voice: "))
 99 | 
100 | if arg_voice == 0:
101 |     voice = lastUsed
102 | else:
103 |     voice = data[arg_voice - 1]
104 | 
105 | if arg_period_pause >= 0:
106 |     ttsText = ttsText.replace(".", ".<break time=\"" + str(ppause) + "s\"/>")
107 |     print("Setting Period Pause To " + str(ppause))
108 | 
109 | if arg_comma_pause >= 0:
110 |     ttsText = ttsText.replace(",", ",<break time=\"" + str(cpause) + "s\"/>")
111 |     print("Setting Comma Pause To " + str(cpause))
112 | 
113 | if arg_line_pause >= 0:
114 |     ttsText = ttsText.replace("\n", "\n<break time=\"" + str(lpause) + "s\"/>")
115 |     print("Setting Newline Pause To " + str(lpause))
116 | 
117 | print("Using Input File (input.txt)")
118 | with open('input.txt', encoding='utf-8') as f:
119 |     ttsText = "".join(f.readlines())
120 | 
121 | #ttsText = ttsText.replace("\r", "").replace("\n", " ")
122 | ttsTextSplit = textwrap.wrap(ttsText, 1500)
123 | ttsTextLen = len(ttsTextSplit)
124 | 
125 | with open("lastused.json", "w") as f:
126 |         f.write(json.dumps(voice))
127 | 
128 | for i in range(ttsTextLen):
129 |     input_text = "<speak><p>" + ttsTextSplit[i] + "</p></speak>"
130 |     params = {
131 |         "globalSpeed": str(arg_speed) + "%",
132 |         "globalVolume": ("+" if arg_volume >= 0 else "") + str(arg_volume) + "dB",
133 |         "chunk": input_text,
134 |         "narrationStyle": "regular",
135 |         "platform": "landing_demo",
136 |         "ssml": input_text,
137 |         "userId": "5pe8l4FrdbczcoHOBkUtp0W37Gh2",
138 |         "voice": voice["value"]
139 |     }
140 |     
141 |     print("Seding request.. [" + str(i + 1) + "/" + str(ttsTextLen) + "]")
142 |     req = requests.post("https://play.ht/api/transcribe", data=params)
143 | 
144 |     filename = "_" + voice["name"] + "-" + str(time.time_ns() / 1000) + "-" + str(i + 1) + ".mp3"
145 |     try:
146 |         response = json.loads(req.text)
147 | 
148 |         head = requests.head(response["file"])
149 |         filesize = head.headers.get('content-length', -1)
150 |         print("Getting file.. [" + str(i + 1) + "/" + str(ttsTextLen) + "] (JSON " + str(round(float(filesize) / 1024, 2)) + " KB)")
151 | 
152 |         with open(filename, "wb") as f:
153 |             f.write(requests.get(response["file"]).content)
154 |     except:
155 |         # Assume we got an audio file
156 |         print("Getting file.. [" + str(i + 1) + "/" + str(ttsTextLen) + "] (MP3? " + str(round(len(req.content) / 1024, 2)) + " KB)")
157 | 
158 |         with open(filename, "wb") as f:
159 |             f.write(req.content)
160 | 
161 |     print("Saved to " + filename)
162 | 


--------------------------------------------------------------------------------