Speech-to-Text for many langages

Google Cloud Build with AI AI APIs
1

Hello,

I’m testing bunch of variety of the Speech-to-Text API to transcribe audio from microphone but I’m going through two issues:

  • It doesn’t reconize others langages like arabic or chinese or at least the transcription shows unknow characters like in this picture below(which is for arabic)

    arabe.PNG
    arabe.PNG622×133 1.92 KB

  • when transcribing audio sentences can be repeated many times

Note: I didn’t see anywhere how to use utf-8 for transcription

How can I fix it I used the code found here https://github.com/googleapis/python-speech/blob/main/samples/microphone/transcribe_streaming_infinite.py

2

I solved the first part of my problem, if anyone is going through the same problem

In windows terminal you can type: chcp 1256 . This allow arabic characters in terminal