deepgeek/CosyVoice
2
0
Fork 0
mirror of https://github.com/FunAudioLLM/CosyVoice.git synced 2026-02-05 18:09:24 +08:00
Code Issues Packages Projects Releases Wiki Activity

update stream code

Browse Source
This commit is contained in:
lyuxiang.lx
2024-07-30 16:11:28 +08:00
parent 02f941d348
commit f4e70e222c
15 changed files with 182 additions and 109 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
cosyvoice/bin/inference.py
View File

@@ -100,10 +100,13 @@ def main():
'flow_prompt_speech_token': speech_token, 'flow_prompt_speech_token_len': speech_token_len,
'prompt_speech_feat': speech_feat, 'prompt_speech_feat_len': speech_feat_len,
'llm_embedding': utt_embedding, 'flow_embedding': utt_embedding}
model_output = model.inference(**model_input)
tts_speeches = []
for model_output in model.inference(**model_input):
tts_speeches.append(model_output['tts_speech'])
tts_speeches = torch.concat(tts_speeches, dim=1)
tts_key = '{}_{}'.format(utts[0], tts_index[0])
tts_fn = os.path.join(args.result_dir, '{}.wav'.format(tts_key))
torchaudio.save(tts_fn, model_output['tts_speech'], sample_rate=22050)
torchaudio.save(tts_fn, tts_speeches, sample_rate=22050)
f.write('{} {}\n'.format(tts_key, tts_fn))
f.flush()
f.close()