neonbjb/tortoise-tts
1
0
Fork 0
mirror of https://github.com/neonbjb/tortoise-tts.git synced 2025-12-06 07:12:00 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge pull request #688 from qtangs/patch-1

Browse source 
Fix for exception on streaming on last chunk - copied from coqui-ai/TTS#3160
This commit is contained in:
manmay nakhashi 2023-11-22 20:56:33 +05:30 committed by GitHub
parent 80f89987a5 8e6c69a407
commit 1e061bc675
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 16 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

13
tortoise/api_fast.py
View file

@ -277,17 +277,28 @@ class TextToSpeech:
settings.update(kwargs) # allow overriding of preset settings with kwargs
for audio_frame in self.tts(text, **settings):
yield audio_frame
# taken from here https://github.com/coqui-ai/TTS/blob/d21f15cc850788f9cdf93dac0321395138665287/TTS/tts/models/xtts.py#L666
# taken from here https://github.com/coqui-ai/TTS/blob/b4c552a112fd4c5f3477f439882eb43c2e2ce85f/TTS/tts/models/xtts.py#L600
def handle_chunks(self, wav_gen, wav_gen_prev, wav_overlap, overlap_len):
"""Handle chunk formatting in streaming mode"""
wav_chunk = wav_gen[:-overlap_len]
if wav_gen_prev is not None:
wav_chunk = wav_gen[(wav_gen_prev.shape[0] - overlap_len) : -overlap_len]
if wav_overlap is not None:
# cross fade the overlap section
if overlap_len > len(wav_chunk):
# wav_chunk is smaller than overlap_len, pass on last wav_gen
if wav_gen_prev is not None:
wav_chunk = wav_gen[(wav_gen_prev.shape[0] - overlap_len):]
else:
# not expecting will hit here as problem happens on last chunk
wav_chunk = wav_gen[-overlap_len:]
return wav_chunk, wav_gen, None
else:
crossfade_wav = wav_chunk[:overlap_len]
crossfade_wav = crossfade_wav * torch.linspace(0.0, 1.0, overlap_len).to(crossfade_wav.device)
wav_chunk[:overlap_len] = wav_overlap * torch.linspace(1.0, 0.0, overlap_len).to(wav_overlap.device)
wav_chunk[:overlap_len] += crossfade_wav
wav_overlap = wav_gen[-overlap_len:]
wav_gen_prev = wav_gen
return wav_chunk, wav_gen_prev, wav_overlap