speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

call streaming speech generation

Browse Source
This commit is contained in:
志浩 2024-09-11 20:37:11 +08:00
parent 397977f4ae
commit d6e401b2da
1 changed files with 5 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
funasr/models/llm_asr/model.py
View File

@ -2918,8 +2918,11 @@ class LLMASRXvecSlotTTS(nn.Module):
):
assert llm_cur_kv_cache is not None
set_all_random_seed(rand_seed)
speech_tokens, mel, wav = self.generate_speech(
response, llm_cur_kv_cache, llm_cur_kv_cache_len, dtype_map[tts_dtype]
# speech_tokens, mel, wav = self.generate_speech(
# response, llm_cur_kv_cache, llm_cur_kv_cache_len, dtype_map[tts_dtype]
# )
speech_tokens, mel, wav = self.simulate_streaming_generate_speech(
preds, llm_cur_kv_cache, llm_cur_kv_cache_len, dtype_map[tts_dtype], tokenizer
)
self.write_mel_wav(kwargs.get("output_dir"), mel, wav, key[0])