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
auto frontend
游雁79b09f1d6779
diff --git a/funasr/models/sense_voice/whisper_lib/model.py b/funasr/models/sense_voice/whisper_lib/model.pyindex 8b3d3ab1c..3d0d6a819 100644--- a/funasr/models/sense_voice/whisper_lib/model.py+++ b/funasr/models/sense_voice/whisper_lib/model.py@@ -27,9 +27,24 @@ class ModelDimensions:n_text_layer: int+# class LayerNorm(nn.LayerNorm):+# def forward(self, x: Tensor) -> Tensor:+# return super().forward(x.float()).type(x.dtype)++class LayerNorm(nn.LayerNorm):- def forward(self, x: Tensor) -> Tensor:- return super().forward(x.float()).type(x.dtype)+ def __init__(self, *args, **kwargs):+ super().__init__(*args, **kwargs)++ def forward(self, input):+ output = F.layer_norm(+ input.float(),+ self.normalized_shape,+ self.weight.float() if self.weight is not None else None,+ self.bias.float() if self.bias is not None else None,+ self.eps,+ )+ return output.type_as(input)class Linear(nn.Linear):