SillyTavern/SillyTavern-Extras
1
0
Fork 0
mirror of https://github.com/SillyTavern/SillyTavern-Extras.git synced 2026-10-16 02:21:52 +00:00
Code Issues 38 Projects Packages Wiki Activity

[GH-ISSUE #34] [Feature] Edge Text to speech #25

New issue
Closed
opened 2026-10-15 18:03:33 +00:00 by bsakaguchi · 5 comments
bsakaguchi commented 2026-10-15 18:03:33 +00:00
Owner
Copy link

Originally created by @FerLuisxd on GitHub (May 26, 2023).
Original GitHub issue: https://github.com/SillyTavern/SillyTavern-Extras/issues/34

Microsoft edge uses some natural sounding voices, would increase the quality of the tts.
There is one pip package that could work!
https://pypi.org/project/edge-tts/

Originally created by @FerLuisxd on GitHub (May 26, 2023). Original GitHub issue: https://github.com/SillyTavern/SillyTavern-Extras/issues/34 Microsoft edge uses some natural sounding voices, would increase the quality of the tts. There is one pip package that could work! https://pypi.org/project/edge-tts/
bsakaguchi commented 2026-10-15 18:03:34 +00:00
Author
Owner
Copy link

@Extraltodeus commented on GitHub (Jun 4, 2023):

I actually opened the issues tab to request that too. These voices are excellent and the service is super reliable while not using any local resource.

I'm using that simple script that I made which can be used with python script_name.py --message "message content" --voice-id 7 --rate-change -10

import warnings
warnings.filterwarnings("ignore")
import sounddevice as sd
import librosa
import edge_tts
import asyncio
import os
import sys
import argparse
import re
from pathlib import Path

parser = argparse.ArgumentParser()
parser.add_argument('--message', type=str, default='Bonjour')
parser.add_argument('--voice-id', type=str, default='7')
parser.add_argument('--rate-change', type=str, default='-7')
args = parser.parse_args()

voix = [
    'fr-BE-CharlineNeural', # 0
    'fr-BE-GerardNeural',   # 1
    'fr-CA-AntoineNeural',  # 2
    'fr-CA-JeanNeural',     # 3
    'fr-CA-SylvieNeural',   # 4
    'fr-CH-ArianeNeural',   # 5
    'fr-CH-FabriceNeural',  # 6
    'fr-FR-DeniseNeural',   # 7
    'fr-FR-EloiseNeural',   # 8
    'fr-FR-HenriNeural'     # 9
]

OUTPUT_FILE = "audio.mp3"

def remove_emojis_and_non_text_chars(text):
    # This regular expression pattern matches emojis and other non-text characters
    emoji_pattern = re.compile(
        "["
        u"\U0001F600-\U0001F64F"  # emoticons
        u"\U0001F300-\U0001F5FF"  # symbols & pictographs
        u"\U0001F680-\U0001F6FF"  # transport & map symbols
        u"\U0001F1E0-\U0001F1FF"  # flags (iOS)
        u"\U00002500-\U00002BEF"  # chinese characters
        u"\U00002702-\U000027B0"
        u"\U00002702-\U000027B0"
        u"\U000024C2-\U0001F251"
        u"\U0001f926-\U0001f937"
        u"\U00010000-\U0010ffff"
        u"\u2640-\u2642"
        u"\u2600-\u2B55"
        u"\u200d"
        u"\u23cf"
        u"\u23e9"
        u"\u231a"
        u"\ufe0f"  # dingbats
        u"\u3030"
        "]+",
        flags=re.UNICODE,
    )
    return emoji_pattern.sub(r"", text)

async def _main(OUTPUT_FILE,TEXT,VOICE=7,RATE=0) -> None:
    sys.stdout = open(os.devnull, 'w')
    if RATE != 0:
        prefix = "+" if RATE > 0 else ""
        RATE = prefix+str(RATE)+"%"
        communicate = edge_tts.Communicate(TEXT,voice=voix[VOICE],rate=RATE)
    else:
        communicate = edge_tts.Communicate(TEXT,voice=voix[VOICE])
    with open(OUTPUT_FILE, "wb") as file:
        async for chunk in communicate.stream():
            if chunk["type"] == "audio":
                file.write(chunk["data"])
            elif chunk["type"] == "WordBoundary":
                print(f"WordBoundary: {chunk}")
    sys.stdout = sys.__stdout__

def tts_voice(TEXT,VOICE=7,RATE=0):
    FOLDER_PATH = "voice_outputs/"
    Path(FOLDER_PATH).mkdir(exist_ok=True)
    OUTPUT_FILE = f"{len(os.listdir(FOLDER_PATH)):04d}_audio.mp3"
    OUTPUT_FILE = os.path.join(FOLDER_PATH,OUTPUT_FILE)
    asyncio.get_event_loop().run_until_complete(_main(OUTPUT_FILE,TEXT,VOICE,RATE))
    audio_data, sample_rate = librosa.load(OUTPUT_FILE)
    sd.play(audio_data, sample_rate)
    sd.wait()

if __name__ == "__main__":
    message = args.message
    message = remove_emojis_and_non_text_chars(message)
    voice_id = int(args.voice_id)
    voice_rate = int(args.rate_change)
    tts_voice(message,voice_id,voice_rate)
<!-- gh-comment-id:1575486762 --> @Extraltodeus commented on GitHub (Jun 4, 2023): I actually opened the issues tab to request that too. These voices are excellent and the service is super reliable while not using any local resource. I'm using that simple script that I made which can be used with python script_name.py --message "message content" --voice-id 7 --rate-change -10 ``` import warnings warnings.filterwarnings("ignore") import sounddevice as sd import librosa import edge_tts import asyncio import os import sys import argparse import re from pathlib import Path parser = argparse.ArgumentParser() parser.add_argument('--message', type=str, default='Bonjour') parser.add_argument('--voice-id', type=str, default='7') parser.add_argument('--rate-change', type=str, default='-7') args = parser.parse_args() voix = [ 'fr-BE-CharlineNeural', # 0 'fr-BE-GerardNeural', # 1 'fr-CA-AntoineNeural', # 2 'fr-CA-JeanNeural', # 3 'fr-CA-SylvieNeural', # 4 'fr-CH-ArianeNeural', # 5 'fr-CH-FabriceNeural', # 6 'fr-FR-DeniseNeural', # 7 'fr-FR-EloiseNeural', # 8 'fr-FR-HenriNeural' # 9 ] OUTPUT_FILE = "audio.mp3" def remove_emojis_and_non_text_chars(text): # This regular expression pattern matches emojis and other non-text characters emoji_pattern = re.compile( "[" u"\U0001F600-\U0001F64F" # emoticons u"\U0001F300-\U0001F5FF" # symbols & pictographs u"\U0001F680-\U0001F6FF" # transport & map symbols u"\U0001F1E0-\U0001F1FF" # flags (iOS) u"\U00002500-\U00002BEF" # chinese characters u"\U00002702-\U000027B0" u"\U00002702-\U000027B0" u"\U000024C2-\U0001F251" u"\U0001f926-\U0001f937" u"\U00010000-\U0010ffff" u"\u2640-\u2642" u"\u2600-\u2B55" u"\u200d" u"\u23cf" u"\u23e9" u"\u231a" u"\ufe0f" # dingbats u"\u3030" "]+", flags=re.UNICODE, ) return emoji_pattern.sub(r"", text) async def _main(OUTPUT_FILE,TEXT,VOICE=7,RATE=0) -> None: sys.stdout = open(os.devnull, 'w') if RATE != 0: prefix = "+" if RATE > 0 else "" RATE = prefix+str(RATE)+"%" communicate = edge_tts.Communicate(TEXT,voice=voix[VOICE],rate=RATE) else: communicate = edge_tts.Communicate(TEXT,voice=voix[VOICE]) with open(OUTPUT_FILE, "wb") as file: async for chunk in communicate.stream(): if chunk["type"] == "audio": file.write(chunk["data"]) elif chunk["type"] == "WordBoundary": print(f"WordBoundary: {chunk}") sys.stdout = sys.__stdout__ def tts_voice(TEXT,VOICE=7,RATE=0): FOLDER_PATH = "voice_outputs/" Path(FOLDER_PATH).mkdir(exist_ok=True) OUTPUT_FILE = f"{len(os.listdir(FOLDER_PATH)):04d}_audio.mp3" OUTPUT_FILE = os.path.join(FOLDER_PATH,OUTPUT_FILE) asyncio.get_event_loop().run_until_complete(_main(OUTPUT_FILE,TEXT,VOICE,RATE)) audio_data, sample_rate = librosa.load(OUTPUT_FILE) sd.play(audio_data, sample_rate) sd.wait() if __name__ == "__main__": message = args.message message = remove_emojis_and_non_text_chars(message) voice_id = int(args.voice_id) voice_rate = int(args.rate_change) tts_voice(message,voice_id,voice_rate) ```
bsakaguchi commented 2026-10-15 18:03:34 +00:00
Author
Owner
Copy link

@Cohee1207 commented on GitHub (Jun 4, 2023):

Did that last night in the main SillyTavern repo, but it didn't work out well on Termux.
Just pushed new API endpoints here, moving to UI plugin support.

<!-- gh-comment-id:1575492033 --> @Cohee1207 commented on GitHub (Jun 4, 2023): Did that last night in the main SillyTavern repo, but it didn't work out well on Termux. Just pushed new API endpoints here, moving to UI plugin support.
bsakaguchi commented 2026-10-15 18:03:34 +00:00
Author
Owner
Copy link

@Extraltodeus commented on GitHub (Jun 4, 2023):

Wow that was a fast answer! :)

Can't wait to try it! (I already copied the files but can't see the option in the UI yet, I failed at reusing your removed files from the last commit on the main repo too lol)

<!-- gh-comment-id:1575532667 --> @Extraltodeus commented on GitHub (Jun 4, 2023): Wow that was a fast answer! :) Can't wait to try it! (I already copied the files but can't see the option in the UI yet, I failed at reusing your removed files from the last commit on the main repo too lol)
bsakaguchi commented 2026-10-15 18:03:34 +00:00
Author
Owner
Copy link

@Cohee1207 commented on GitHub (Jun 4, 2023):

Added to dev branch of ST.

<!-- gh-comment-id:1575533689 --> @Cohee1207 commented on GitHub (Jun 4, 2023): Added to dev branch of ST.
bsakaguchi commented 2026-10-15 18:03:34 +00:00
Author
Owner
Copy link

@Extraltodeus commented on GitHub (Jun 4, 2023):

Oh thank you! It works :)

<!-- gh-comment-id:1575539448 --> @Extraltodeus commented on GitHub (Jun 4, 2023): Oh thank you! It works :)
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
No results found.
Labels
Clear labels   
bug

   
duplicate

   
planned

   
pull-request
Mirrored from GitHub Pull Request

  
under consideration
No labels
bug
duplicate
planned
pull-request
under consideration
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
SillyTavern/SillyTavern-Extras#25
No description provided.