Groq

Groq are a super fast and efficient AI provider. The LPU™ Inference Engine by Groq is a hardware and software platform that delivers exceptional compute speed, quality, and energy efficiency. Orate supports Groq's transcription services.

Setup

The Groq provider is available by default in Orate. To import it, you can use the following code:

import { groq } from 'orate/groq';

Configuration

The Groq provider looks for the GROQ_API_KEY environment variable. This variable is required for the provider to work. Simply add the following to your .env file:

GROQ_API_KEY="your_api_key"

Usage

The Groq provider provides a single interface for all of Groq's transcription services.

Speech to Text

The Groq provider provides a stt function that allows you to create a speech-to-text transcription function using Groq Whisper. By default, the stt function uses the whisper-large-v3 model.

import { transcribe } from 'orate';
import { groq } from 'orate/groq';
 
const text = await transcribe({
  model: groq.stt(),
  audio: someArrayBuffer,
});

You can specify the model to use by passing it as an argument to the stt function.

const text = await transcribe({
  model: groq.stt('whisper-large-v3'),
  audio: someArrayBuffer,
});

You can also specify specific Groq properties by passing them as an argument to the stt function.

const text = await transcribe({
  model: groq.stt('whisper-large-v3', {
    temperature: 0.5,
  }),
  audio: someArrayBuffer,
});