deepgeek/CosyVoice
2
0
Fork 0
mirror of https://github.com/FunAudioLLM/CosyVoice.git synced 2026-02-05 01:49:25 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix padding

Browse Source
This commit is contained in:
lyuxiang.lx
2026-01-07 07:12:52 +00:00
parent f97d50d559
commit 7fdd80dc64
1 changed files with 1 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
cosyvoice/llm/llm.py
View File

@@ -314,6 +314,7 @@ class Qwen2LM(TransformerLM):
else: else:
instruct_token = [torch.empty(0).to(text_token[0])] * len(text_token) instruct_token = [torch.empty(0).to(text_token[0])] * len(text_token)
instruct_token_emb = [torch.empty(0, 896).to(text_token_emb[0])] * len(text_token) instruct_token_emb = [torch.empty(0, 896).to(text_token_emb[0])] * len(text_token)
instruct_token_len = torch.zeros(len(text_token)).to(text_token_len)
for i in range(len(text_token)): for i in range(len(text_token)):
# bistream sequence # bistream sequence
if random.random() < 0.5 and speech_token_len[i] / text_token_len[i] > self.mix_ratio[1] / self.mix_ratio[0]: if random.random() < 0.5 and speech_token_len[i] / text_token_len[i] > self.mix_ratio[1] / self.mix_ratio[0]: