# ElevenLabs Scribe v2

Writes down what is said in a clip or a sound, with the time each word starts and ends.

Model id: `scribe-v2`
Type: stt
Inputs: Audio
Updated: Oct 29, 2026

## Pricing

| Setting | MiniAPI | fal | Difference |
| --- | ---: | ---: | ---: |
| any | $0.0058 / minute of audio | $0.008 | -28% |

You are not charged for a job that fails.

## Parameters

| Name | Required | Values | Default | Description |
| --- | --- | --- | --- | --- |
| `audio_url` | Yes | https URL |  | The sound to use. |
| `language_code` | No | text |  | Language code of the audio |
| `tag_audio_events` | No | true, false | true | Tag audio events like laughter, applause, etc. |
| `diarize` | No | true, false | true | Whether to annotate who is speaking |
| `keyterms` | No |  |  | Words or sentences to bias the model towards transcribing. Up to 100 keyterms, max 50 characters each. Adds 30% premium over base transcription price. |

Page: https://api.minicut.ai/models/scribe-v2
API reference: https://api.minicut.ai/docs/api