Skip to content

Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.

bugfix seg_dict_file

游雁e02f361de651
diff --git a/funasr/tokenizer/char_tokenizer.py b/funasr/tokenizer/char_tokenizer.py
index 28b4f407c..2efc0b056 100644
--- a/funasr/tokenizer/char_tokenizer.py
+++ b/funasr/tokenizer/char_tokenizer.py
@@ -36,6 +36,7 @@ class CharTokenizer(BaseTokenizer):
self.remove_non_linguistic_symbols = remove_non_linguistic_symbols
self.split_with_space = split_with_space
self.seg_dict = None
+ seg_dict = seg_dict if seg_dict is not None else kwargs.get("seg_dict_file", None)
if seg_dict is not None:
self.seg_dict = load_seg_dict(seg_dict)