I need help getting the transcribed text from Google Cloud Speech-to-Text asynchronous operations in Make, as the result isn't available after using "Get the State."

How To
,
1

Hello Community,

I’m setting up a Make flow to transcribe long audio files (over 10MB) using Google Cloud Speech-to-Text’s asynchronous transcription.

My current flow sequence is:

  1. Google Drive (Download audio)

  2. CloudConvert (to WAV)

  3. Google Cloud Storage (Upload WAV)

  4. Google Cloud Speech - Start Asynchronous Speech Recognition (Module 5)

  5. Google Cloud Speech - Get the State of Asynchronous Speech Recognition (Module 6)

  6. Google Docs (Create Document)

The Problem:

  • Module 6 (Get the State...) executes successfully (status is DONE), but its Output only contains the Metadata (including the operation ID), not the actual transcribed text.

  • My list of Google Cloud Speech modules does not include a “Get the Result of Asynchronous Speech Recognition” module.

  • The “Start Asynchronous…” module also does not have a visible Output URI field in the advanced settings to specify where to save the JSON result file.

My Question is:

Since the dedicated “Get the Result” module is missing, and the Output URI field is not visible, where does Make/Google Cloud automatically save the final JSON transcription result? And what is the correct sequence of GCS/JSON modules I should use to download and parse that JSON to retrieve the transcribed text?