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
fix: add repetition_penalty=1.3 to streaming service (fixes #2930 streaming repetition)
游雁91905fd549d4
diff --git a/examples/industrial_data_pretraining/fun_asr_nano/serve_realtime_ws.py b/examples/industrial_data_pretraining/fun_asr_nano/serve_realtime_ws.pyindex 5c58bc5ab..b52a1dd1d 100644--- a/examples/industrial_data_pretraining/fun_asr_nano/serve_realtime_ws.py+++ b/examples/industrial_data_pretraining/fun_asr_nano/serve_realtime_ws.py@@ -318,6 +318,7 @@ class RealtimeASRSession:hotwords=self.asr_kwargs.get("hotwords"),language=self.asr_kwargs.get("language"),max_new_tokens=200,+ repetition_penalty=1.3,)text = results[0]["text"] if results else ""text = _clean_asr_text(text)@@ -358,6 +359,7 @@ class RealtimeASRSession:hotwords=self.asr_kwargs.get("hotwords"),language=self.asr_kwargs.get("language"),max_new_tokens=512,+ repetition_penalty=1.3,)text = results[0]["text"] if results else ""text = _clean_asr_text(text)diff --git a/funasr/version.txt b/funasr/version.txtindex 95b25aee2..3336003dc 100644--- a/funasr/version.txt+++ b/funasr/version.txt@@ -1 +1 @@-1.3.6+1.3.7