Skip to content

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

Fix invalid escape sequence '\w' in RegEx due to 'r' missing (#1967)

Ziyao Wang9f04fd130c84
diff --git a/funasr/models/llm_asr/model.py b/funasr/models/llm_asr/model.py
index b4d9e7ce7..fa1d2c3c2 100644
--- a/funasr/models/llm_asr/model.py
+++ b/funasr/models/llm_asr/model.py
@@ -812,7 +812,7 @@ class LLMASR2(nn.Module):
ibest_writer = self.writer[f"{0 + 1}best_recog"]
results = []
- response_clean = re.sub("[^\w\s\u3000\u4e00-\u9fff]+", "", response)
+ response_clean = re.sub(r"[^\w\s\u3000\u4e00-\u9fff]+", "", response)
result_i = {"key": key[0], "text": response, "text_tn": response_clean, "label": label}
if loss is not None:
result_i["loss"] = loss
@@ -1395,7 +1395,7 @@ class LLMASR4(nn.Module):
ibest_writer = self.writer[f"{0 + 1}best_recog"]
results = []
- response_clean = re.sub("[^\w\s\u3000\u4e00-\u9fff]+", "", response)
+ response_clean = re.sub(r"[^\w\s\u3000\u4e00-\u9fff]+", "", response)
result_i = {"key": key[0], "text": response, "text_tn": response_clean, "label": label}
if loss is not None:
result_i["loss"] = loss