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60b3c42d6d3d
diff --git a/funasr/models/transformer/attention.py b/funasr/models/transformer/attention.pyindex e4add5cf8..6e6f75411 100644--- a/funasr/models/transformer/attention.py+++ b/funasr/models/transformer/attention.py@@ -82,7 +82,10 @@ class MultiHeadedAttention(nn.Module):n_batch = value.size(0)if mask is not None:mask = mask.unsqueeze(1).eq(0) # (batch, 1, *, time2)- min_value = float(numpy.finfo(torch.tensor(0, dtype=scores.dtype).numpy().dtype).min)++ min_value = -float(+ "inf"+ ) # min_value = float(np.finfo(torch.tensor(0, dtype=qk.dtype).numpy().dtype).min)scores = scores.masked_fill(mask, min_value)self.attn = torch.softmax(scores, dim=-1).masked_fill(mask, 0.0