Speech Isolation

Orate provides a simple and unified API for isolating speech from the audio using various AI providers. It's a very handy function for removing background noise from chaotic audio. The speech isolation functionality is accessible through the isolate function.

Usage

The basic usage involves importing the isolate function from Orate and your chosen provider. For example, to use ElevenLabs' speech isolation model, you can import the ElevenLabs provider and use its isl function:

import { isolate } from 'orate';
import { elevenlabs } from 'orate/elevenlabs';
 
const transcription = await isolate({
  model: elevenlabs.isl(),
  audio,
});

model

The model parameter is a function that returns a model instance. The model instance is a function that takes a prompt and returns a promise that resolves to the speech of the prompt.

const model = elevenlabs.isl();

audio

The audio parameter is a File object that contains the audio to be processed.

const audio = new File([], 'test.mp3', { type: 'audio/mp3' });