SpeechRecognition

DeepSpeech
#1

Hello,
Does anyone know if it is possible to use the SpeechRecognition API
to receive audio through streaming instead of receiving through the microphone?

#2

What is this API ? Please elaborate.

#3

Hello @lissyx.
https://developer.mozilla.org/en-US/docs/Web/API/SpeechRecognition .

#4

The W3C SpeechRecognition API is not supposed to send you audio …

#5

So @lissyx, But the W3C SpeechRecognition is able to receive the audio input from the computer’s microphone (This I have already tested and works well).
My idea was to do the same procedure, receiving from a streaming channel.
To be more specific:
I want to do a project and develop something that takes audio from a specific program on a Web radio and transcribes it for a deaf person to read.

#6

You want to feed some third-party audio into SpeechRecognition API ?

  • This is obviously not a deepspeech-related question
  • This is obviously not something you can do without changing the API implementation
  • We have no deepspeech usage in the SpeechRecognition API.
#7

If you got the audio somehow, you use one of the examples to transcribe it with DeepSpeech. How you get the audio will unfortunately be your problem.

1 Like
#8

@othiele It seems to me that this will satisfy.
I will read the documentation.
Thank you for now