Skip to content

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

bug fix

shixian.shi7ab24f4e7465
diff --git a/funasr/auto/auto_model.py b/funasr/auto/auto_model.py
index 6d74f81e6..f724650f3 100644
--- a/funasr/auto/auto_model.py
+++ b/funasr/auto/auto_model.py
@@ -377,7 +377,7 @@ class AutoModel:
result[k] = restored_data[j][k]
else:
result[k] = torch.cat([result[k], restored_data[j][k]], dim=0)
- elif k == 'text':
+ elif k == 'raw_text':
if k not in result:
result[k] = restored_data[j][k]
else: