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
cer
游雁1ef81172139e
diff --cc funasr/models/llm_asr_nar/model.pyindex f17034930,6a4eccefe..db81c47d9--- a/funasr/models/llm_asr_nar/model.py+++ b/funasr/models/llm_asr_nar/model.py@@@ -315,8 -315,7 +315,10 @@@ class LLMASRNAR(nn.Module)model_outputs = self.llm(inputs_embeds=inputs_embeds, attention_mask=attention_mask, labels=None)preds = torch.argmax(model_outputs.logits, -1)text = tokenizer.batch_decode(preds, add_special_tokens=False, skip_special_tokens=True)- text = text[0].split(': \n')[-1]+++ text = text[0].split(': ')[-1]+ text = text.strip()++# preds = torch.argmax(model_outputs.logits, -1)ibest_writer = None