I want user input as Audio and use an STT engine which is not there in appFoundry

1

I want to take user input as audio and transcript that audio in an Opensource transcriptor whose accuracy is far better than Google transcriptor.
We are thinking like, if we can get the audio in any format, we can send that audio to our 3rd party opensource API, which will give us the transcripted text.

For example: we want to get the User address via Audio. So, google transcription doesn't work well in that case. We have found a service which transcript with more accuracy than google.

Then, based on the address we want to serve the user.

Is there any provision in genesys to achieve this ?

2

GET /api/v2/conversations/{conversationId}/recordings

3

Hi Ebenezer, We want the user input realtime so that we can transcript the audio and based on it, we can serve the user.

For example: we want to get the User address via Audio. So, google transcription doesn't work well in that case. We have found a service which transcript with more accuracy than google.

Then, based on the address we want to serve the user.
It is real-time.

4

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.