Mprove speech to text deep speech

DeepSpeech
#1

Hello . When I downloaded deepspeech and run it on Windows, it unfortunately turned bad speech into text.

For example, when I say:
HI ----> i
you are ->you are

hello -> halow

How can I increase the accuracy or efficiency of speech to text conversion?

I just want to use only the Deep Speech model and I do not want to teach on any datasets? Is there a way or not?

When I say a word through a microphone, do I already have to make certain settings in the Windows environment?..

#2

The current model is bad with accents other than US English. Try a snippet from youtube to test the model if you think this might be the cause. Otherwise use wav files to make debugging easier.

1 Like
#3

thank you . Do I need to make any special settings for the microphone in Windows?

For example, how many bits and how many channels and how many hertz?

#4

Please make the effort to search before you post, this is a really common question.