Text to speech in Make.com: one HTTP module, any voice
Make (formerly Integromat) connects apps into automated scenarios: a new row, a form, a message or an RSS item can start a chain of steps. When one of those steps should produce speech, an HTTP request to a text to speech API is all it takes.
This guide shows how to voice text in a Make scenario with the SuperClonica speech API and pass the audio on to the next module.
Get a key
Sign in to SuperClonica with Google or GitHub and create an API key in the Speech API section of your workspace. The free plan gives a minute of audio a day for testing; the $5 Starter plan gives up to 216 hours a month.
Set up the HTTP module
Add the HTTP app’s “Make a request” module after the step that provides the text and fill it in:
- URL: https://superclonica.com/api/v1/speech, method POST.
- Headers: Authorization with the value Bearer and your key; Content-Type: application/json.
- Body type Raw, content type JSON, request content: {"text": "{{the text field}}", "voice": "F1", "language": "en", "format": "mp3"}.
- Leave “Parse response” off: the answer is the audio file itself, which the module passes on as binary data.
Save or send the audio
Map the module’s Data output into the next module that takes a file: Google Drive “Upload a file”, Dropbox, an email attachment or a Telegram bot “Send an audio”. Give the file a name ending in .mp3. If you want each run to be safe to repeat, add an Idempotency-Key header with the ID of the trigger’s record.
Errors and limits
Turn on the module’s option to evaluate errors by status code, then add an error handler: 429 means the plan’s 10-minute window is used up and 503 that the queue is full; both carry a Retry-After header with the seconds to wait. One request takes up to 20,000 characters. Through the API the text goes to SuperClonica servers; for private texts run the engine on your own machine instead.