一个可以将输入的视频文件提取出其中的音频，并对识别音频中的语音并输出到文本框里的应用程序，现在准确率还比较低，望牛人指教如何提高准确率-An input video file can be extracted out of one of the audio, and audio of the speech recognition and output to a text box applications, is now relative