When making the speech recognition, in order to improve the recognition rate, all training first, because I am a need to make the speech recognition engine familiar voice in the audio files, such as news broadcast hosted by voice, rather than use a microphone to familiar with someone's voice, so I hope can wav files for training, want to give a great god help action

CodePudding user response:

Look at the previous post, one way is to install a virtual sound card software, the wav file output to the virtual sound card, and then as an input to the SAPI
If the trouble, the simplest way is to use speaker playing wav out, then the opposite which microphone to identify, this approach any format can be, not limited to wav format, you can also directly to the training of the television, effect better

CodePudding user response:

Hello, I'm sorry, the first time to do something, take this thing to delay for several days, I saw your reply, thank you very much, but the installation method of virtual sound card software I am not quite understand, can you say about it? Thank you very much

CodePudding user response:

This I also did not install
But the search on the net, there are a lot of virtual sound card software, find there should be proper

CodePudding user response:

Ok thank you very much