diff --git a/SpeechModel251.py b/SpeechModel251.py index f967cae..fbc9c9d 100644 --- a/SpeechModel251.py +++ b/SpeechModel251.py @@ -324,7 +324,7 @@ class ModelSpeech(): # 语音模型类 #print('r', r) r1 = r[0][0].eval(session=tf.compat.v1.Session()) - tf.compat.v1.reset_default_graph() + #tf.compat.v1.reset_default_graph() return r1[0] def RecognizeSpeech(self, wavsignal, fs):