bugfix: audio_len should be 1D, no 0D, which will raise list index out (

#3490)

of range error in the following decode process

Co-authored-by: Luzhenhui <luzhenhui@mqsz.com>

paddlespeech/cli/asr/infer.py

@@ -274,7 +274,7 @@ def preprocess(self, model_type: str, input: Union[str, os.PathLike]):
             # fbank
             audio = preprocessing(audio, **preprocess_args)
 
-            audio_len = paddle.to_tensor(audio.shape[0]).unsqueeze(axis=0)
+            audio_len = paddle.to_tensor([audio.shape[0]]).unsqueeze(axis=0)
             audio = paddle.to_tensor(audio, dtype='float32').unsqueeze(axis=0)
 
             self._inputs["audio"] = audio