Trying to run a scenario using Google Cloud Speech but I get a strange error: 400 Invalid value at ‘audio’ (oneof), oneof field ‘audio_source’ is already set. Cannot set ‘content’ and I’m not able to figure out how to solve it.
The audio file is stored in Google Cloud Storage as it should be according to the documentation but still, this error appears.
The same problem. Or I get empty metadata, otherwise the same issue if use synchronous speech recognition.
I finally manage to get it working. Attached please find the exact config of my Google STT module. “The audio file” is the mp3 file I get from an email attachment and “data”. It is important to know the mp3 file “Sample rate in Hertz” and the number of audio channels. The “Original MIME type” I get after uploading the file to Google Drive - it’s available as the file metadata.
Heya @surmacki 
I just wanted to quickly jump in and say thanks a lot for stepping back in here and sharing your solution with us. It’s great to see that you’re keeping the community in mind 
Just FYI: I marked your reply as a solution to keep the space tidy and to make it easier for folks who are looking for similar information in the future. 
I have the same issue!
And the docs are not helpful, since it seems the bug is in the hidden fields sent to google.
The “google.cloud.speech.v1p1beta1.Speech.Recognize” action is invoked, and responding with “400” status.
I tried using the “async” Recognize.
It works, but don’t understand how to fetch the response text.
The docs don’t address that, and I don’t understand where do I fetch the transpiled text from.
But the response is errored.
