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
decoding
游雁765e6371bbfb
diff --git a/funasr/models/llm_asr/model.py b/funasr/models/llm_asr/model.pyindex e7dd11ea3..f5d6a82ab 100644--- a/funasr/models/llm_asr/model.py+++ b/funasr/models/llm_asr/model.py@@ -407,7 +407,7 @@ class LLMASR2(nn.Module):audio_encoder = encoder_class(input_size=input_size, **audio_encoder_conf)audio_encoder_output_size = audio_encoder.output_size()freeze = audio_encoder_conf.get("freeze", True)- freeze_layer_num = audio_encoder_conf.get("freeze_layer_num", -1)+ freeze_layer_num = int(audio_encoder_conf.get("freeze_layer_num", -1))if freeze_layer_num > 0:freeze_layer_num = range(freeze_layer_num)else:diff --git a/funasr/train_utils/load_pretrained_model.py b/funasr/train_utils/load_pretrained_model.pyindex cb2d87e87..6bada5f85 100644--- a/funasr/train_utils/load_pretrained_model.py+++ b/funasr/train_utils/load_pretrained_model.py@@ -60,7 +60,7 @@ def load_pretrained_model(if excludes is not None:for k_ex in excludes:if k.startswith(k_ex):- logging.info(f"key: {{k}} matching: {k_ex}, excluded")+ logging.info(f"key: {k} matching: {k_ex}, excluded")excludes_flag = Truebreakif excludes_flag: