Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Python asraudiochineseemotion-recognitionfunasrmcp-servermultilingual-asropenai-compatible-apiparaformerpunctuationpytorchreal-time-asrspeaker-diarizationspeech-recognitionspeech-to-textstreaming-asrtranscriptionvllmvoice-activity-detectionwhisper-alternative
修复断句之间时间戳bug (#1863)
wuhongshengd8c1b46daf4f
diff --git a/funasr/utils/timestamp_tools.py b/funasr/utils/timestamp_tools.pyindex 6abebe165..4fc2d15c5 100644--- a/funasr/utils/timestamp_tools.py+++ b/funasr/utils/timestamp_tools.py@@ -156,6 +156,7 @@ def timestamp_sentence(punc_id = int(punc_id) if punc_id is not None else 1sentence_end = timestamp[1] if timestamp is not None else sentence_end+ sentence_start = timestamp[0] if timestamp is not None else sentence_startsentence_text_seg = (sentence_text_seg[:-1] if sentence_text_seg[-1] == " " else sentence_text_seg)@@ -183,7 +184,6 @@ def timestamp_sentence(sentence_text = ""sentence_text_seg = ""ts_list = []- sentence_start = sentence_endreturn res