Groq automation with Bizomate
A voice note from a driver, a text message from a customer: someone has to listen or read before anything happens. Bizomate asks Groq for each item instead — an open model, such as Llama, answers with text or named fields, and Whisper turns a recording into text. It runs on your company's own Groq key, and every run's record shows the tokens or seconds of audio used.
At a glance
| Label | Value |
|---|---|
| Plans | Every plan |
| Connect | A Groq API key |
| Starts runs | No trigger — starts from another app, a schedule or a webhook |
| Kind | Connected app · AI |
What a workflow can do with it
Ask a fast open model, get text or fields
Choose Fields for named answers the next steps use one by one, or Text. The Model list is read from Groq with your key.
Turns a recording into text with Whisper, with times
A recording from an earlier step — MP3, M4A, WAV, OGG or WebM, up to 25 MB. Language starts at Detect; Prompt takes words the recording uses that a model might mishear, such as Northwind Logistics. The text comes back in timed parts.
Questions about Groq
Who pays for what Groq does?
Your company does, on its own Groq account. Bizomate does not pay for or resell AI use; each run's output shows the tokens or seconds of audio it used.
Which recordings can Transcribe audio take?
A file from an earlier step — an email attachment, a form upload — in MP3, M4A, WAV, OGG or WebM, up to 25 MB.
How do I get names and order numbers transcribed correctly?
Put them in Prompt — the words the recording uses that a model might mishear, such as a company name or an order number. Choosing the Language helps with short recordings too.
What does Bizomate send to Groq?
Only what the step is given — its instructions and prompt, or the recording — under your account, when the step runs. Your account's terms with Groq apply to it.