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
bugfix
游雁199f6e630af9
diff --git a/examples/industrial_data_pretraining/paraformer/finetune.sh b/examples/industrial_data_pretraining/paraformer/finetune.shindex fe511ff25..9467a0b76 100644--- a/examples/industrial_data_pretraining/paraformer/finetune.sh+++ b/examples/industrial_data_pretraining/paraformer/finetune.sh@@ -10,7 +10,6 @@ gpu_num=$(echo $CUDA_VISIBLE_DEVICES | awk -F "," '{print NF}')## option 1, download model automaticallymodel_name_or_model_dir="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"-model_name_or_model_dir="iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"## option 2, download model by git#local_path_root=${workspace}/modelscope_modelsdiff --git a/funasr/models/sense_voice/model.py b/funasr/models/sense_voice/model.pyindex 242664bee..4ee2fa55c 100644--- a/funasr/models/sense_voice/model.py+++ b/funasr/models/sense_voice/model.py@@ -96,7 +96,7 @@ class SenseVoice(nn.Module):options = whisper.DecodingOptions(**DecodingOptions)result = whisper.decode(self.model, speech, options)- text = f"{result.text}\n"+ text = f"{result.text}"results = []result_i = {"key": key[0], "text": text}