Dev gzf exp (#1651)

* sensevoice finetune * sensevoice finetune * sensevoice finetune * sensevoice finetune * sensevoice finetune * sensevoice finetune * sensevoice finetune * sensevoice finetune * sensevoice finetune * sensevoice finetune * bugfix * update with main (#1631) * update seaco finetune * v1.0.24 --------- Co-authored-by: 维石 <shixian.shi@alibaba-inc.com> * sensevoice * sensevoice * sensevoice * update with main (#1638) * update seaco finetune * v1.0.24 * update rwkv template --------- Co-authored-by: 维石 <shixian.shi@alibaba-inc.com> * sensevoice * sensevoice * sensevoice * sensevoice * sensevoice * sensevoice * sensevoice * sensevoice * sensevoice * sensevoice * sensevoice * sensevoice * sensevoice * sensevoice * sensevoice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice * sense voice --------- Co-authored-by: 维石 <shixian.shi@alibaba-inc.com>
2025-09-15 14:48:36 +08:00 · 2024-04-23 20:13:44 +08:00 · 2024-04-23 20:13:44 +08:00 · 26d642bfdf
commit 26d642bfdf
parent 61d631fb5b
4 changed files with 2 additions and 109094 deletions
--- a/funasr/models/sense_voice/model.py
+++ b/funasr/models/sense_voice/model.py
@ -209,7 +209,7 @@ class SenseVoice(nn.Module):
        language = None if language == "auto" else language
        DecodingOptions["language"] = language

-        DecodingOptions["vocab_path"] = kwargs.get("vocab_path", None)
+        DecodingOptions["vocab_path"] = kwargs["tokenizer_conf"].get("vocab_path", None)
        
        
        if "without_timestamps" not in DecodingOptions:
@ -421,7 +421,7 @@ class SenseVoiceRWKV(nn.Module):
        language = None if language == "auto" else language
        DecodingOptions["language"] = language
        
-        DecodingOptions["vocab_path"] = kwargs.get("vocab_path", None)
+        DecodingOptions["vocab_path"] = kwargs["tokenizer_conf"].get("vocab_path", None)
        
        if "without_timestamps" not in DecodingOptions:
            DecodingOptions["without_timestamps"] = True
--- a/funasr/models/sense_voice/whisper_lib/assets/gpt2.tiktoken
+++ b/funasr/models/sense_voice/whisper_lib/assets/gpt2.tiktoken
--- a/funasr/models/sense_voice/whisper_lib/assets/mel_filters.npz
+++ b/funasr/models/sense_voice/whisper_lib/assets/mel_filters.npz
--- a/funasr/models/sense_voice/whisper_lib/assets/multilingual.tiktoken
+++ b/funasr/models/sense_voice/whisper_lib/assets/multilingual.tiktoken