whisper-medium_timestamped

import { pipeline } from '@huggingface/transformers';

const transcriber = await pipeline(
  'automatic-speech-recognition',
  'neonwatty/whisper-medium_timestamped'
);

const result = await transcriber(audioUrl, {
  return_timestamps: 'word',
  chunk_length_s: 30,
  stride_length_s: 5,
});

console.log(result);
// { text: "Hello world", chunks: [{ text: "Hello", timestamp: [0.0, 0.5] }, ...] }