Skip to content

Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.

[fix] Fix a bug in seaco_paraformer model "inference" function (#1639)

querryton01df8f330ccc
diff --git a/funasr/models/seaco_paraformer/model.py b/funasr/models/seaco_paraformer/model.py
index 4778a7ec7..b28de94f7 100644
--- a/funasr/models/seaco_paraformer/model.py
+++ b/funasr/models/seaco_paraformer/model.py
@@ -350,7 +350,7 @@ class SeacoParaformer(BiCifParaformer, Paraformer):
pre_acoustic_embeds, pre_token_length = predictor_outs[0], predictor_outs[1]
pre_token_length = pre_token_length.round().long()
if torch.max(pre_token_length) < 1:
- return []
+ return ([],)
decoder_out = self._seaco_decode_with_ASF(encoder_out,
encoder_out_lens,