Compare accuracy across AWS, Bing and Google API to convert audio to text.
The audio is a low quality recording of an interview of 45 mins.
Text has been converted to audio. Need to calculate metrics to quantify errors.
-
Set up an account in Google cloud and created a new project.
-
Set the 'GOOGLE_APPLICATION_CREDENTIALS' environment variable.
-
Installed Google client library (Google-Cloud-Speech)
-
Trying to run Python script to make audio transcription error (Getting an error with the request)