Skip to content
Nappai Documentation

Groq Whisper STT

Groq Whisper STT is one of the components that act as the hearing of a voice agent: it turns what the caller says into text so the agent can understand it. It uses Whisper speech recognition on the Groq service. It transcribes each full sentence once the caller stops speaking.

It plugs into the STT connector of the Voice Agent node. It holds the speech recognition settings the agent uses on every call.

When to use it

  • The agent’s model is already on Groq and you want to use the same account for hearing.
  • You prefer accuracy on full sentences over the agent starting to think while the caller speaks.
  • Callers may speak different languages and you want the agent to recognise them.

Configuration

You need a Groq API credential with your Groq account key. See Credentials.

FieldWhat it does
Credential (required)The Groq API credential used to connect to the service.
LanguageCode of the language the caller is expected to speak, for example es or en. Default: es.
Detect LanguageWhen on, the language of each sentence is recognised. If the result is a language not in Supported Languages, the sentence is transcribed again in the Language you set. Off by default.
Modelturbo (default) is the fastest and cheapest. large is the full model, with better results in Spanish.

Advanced options

FieldWhat it does
Supported LanguagesLanguages accepted as the result of detection. Default: es and en. Only used when Detect Language is on.

Output

The STT output holds the speech recognition settings. It connects to the STT connector of the Voice Agent node, or to Fallback STT when you turn on Fallback Engines and want it as a backup. The editor will not let you plug it into a TTS connector by mistake.

Example

A law firm wants accurate transcripts of what its clients dictate over the phone.

  1. Add Groq Whisper STT and connect it to the STT connector of the Voice Agent.
  2. In Credential, choose the Groq API credential.
  3. Choose large in Model and leave Language as es.
  4. Test the agent in the playground by dictating a name and an address.

Good practices

  • Groq bills the usage to your own Groq account. Check your usage there.
  • The differences between local and cloud engines are explained in Quality, noise and latency.