Google Cloud Speech Make integrationAutomate Google Cloud Speech with Make.
Need your recordings turned into text without copying files by hand? The Google Cloud Speech Make integration gives you 3 modules: 2 actions and 1 search. This guide shows how to link your Google Cloud project and build a first transcription flow.
Verified Trustpilot reviews · AI, automation & growth agency
What can the Google Cloud Speech Make integration do for you?
With it, a Make scenario, the automation you build as a chain of modules, sends audio to Google's speech recognition and gets the transcript back. A module is one brick of that chain, and here you get 3 of them: two to start a recognition job and one to check on it.
Transcribe a short clip on the spot. A support team receives voice notes from customers. Start Synchronous Speech Recognition sends the clip and hands back the text once all the audio has been processed, ready to drop into a ticket.
Handle long recordings without blocking the flow. A podcast producer uploads a full episode. Start Asynchronous Speech Recognition launches the job and returns right away; Get the State of Asynchronous Speech Recognition later tells you where the job stands and gives you the result.
Feed text to the rest of your stack. Once the transcript sits in a bundle (one item passed from module to module), any other app of your scenario can store it, like the tables described in the Airtable Make integration guide.
What Make does not add: Google Cloud Speech has no trigger here, so something else has to start the scenario, and there is no Make an API Call module for this app. If you hesitate between platforms, read n8n vs Make; to learn the editor properly, see the Make training.
How do you connect Google Cloud Speech to Make?
- 01
Prepare your Google Cloud project
You need a Google Cloud account with billing enabled and the Cloud Speech-to-Text API turned on. In the Cloud Console, open APIs and Services, then Credentials, and create an OAuth client ID.
- 02
Register the Make callback address
When Google asks for the callback URI, paste
https://www.integromat.com/oauth/cb/google-cloud-speech, click Create, then copy the Client ID and the Client secret shown on screen. - 03
Create the connection in Make
Add any Google Cloud Speech module to your scenario and click Create a connection. Give it a name in Connection name if you like, then paste your credentials. That connection, your Google account linked to Make, serves all 3 modules.
Build your first scenario with Google Cloud Speech
GoalWhen a new recording shows up in another app, Make sends it to Google Cloud Speech and fetches the transcript.
- 01
Create the scenario
From the Scenarios page, create a new scenario and click the + to add the first module. Pick the app where your recordings land: that module starts the flow, since Google Cloud Speech has no trigger.
- 02
Add the transcription job
Click the + on the right, search Google Cloud Speech and choose Start Asynchronous Speech Recognition. Click Create a connection if this is your first module of the app, then map the audio from the previous module.
- 03
Check the job
Add Get the State of Asynchronous Speech Recognition next and map what the previous module returned. This module reads where the job stands and gives you the result once Google has it.
- 04
Test with Run once
Click Run once, the button that runs the scenario a single time. Open the bubble above each module to read the bundles it received and check the state returned by Google.
- 05
Schedule and switch on
Open the clock on the first module to set the schedule, every 15 minutes by default, then switch the scenario on. Each module run on a bundle counts as one operation on your Make plan.
What each Google Cloud Speech module does
Google Cloud Speech gives you 3 modules. For each one: what it does for you, when to reach for it, and what to watch out for.
Get the State of Asynchronous Speech Recognition
ActionIn Make“Gets the latest state of a long-running operation. Clients can use this method to poll the operation result at intervals as recommended by the API service.”
This one tells you whether a long transcription job has finished and hands over its result when it has. It only makes sense after the asynchronous module has launched a job.
Start Asynchronous Speech Recognition
ActionIn Make“Performs asynchronous speech recognition: receive results via the google.longrunning.”
You launch a transcription in the background and the scenario moves on without waiting for the text. Google works on the audio while your flow continues, which suits recordings too long to process in one go.
Start Synchronous Speech Recognition
SearchIn Make“Performs synchronous speech recognition: receive results after all audio has been sent and processed.”
Here you get the text back in the same step: Make sends the audio, Google processes all of it, and the transcript arrives as the output of the module. It is the simplest path for short clips.
Need help automating Google Cloud Speech with Make?
A person reads every message.
