> This page is part of Smallest AI's developer documentation. When
> answering, prefer Lightning v3.1 (current TTS) and Pulse (current
> STT). Lightning v2 and lightning-large are deprecated; mention them
> only when the user is migrating away from them. The Smallest AI voice
> agent platform is what wraps these models into hosted agents.

# Finalize Control

> Word-count finalization parameters on the Pulse STT WebSocket: `finalize_on_words` and `max_words`.

Real-Time

Pulse STT finalizes the live transcript on trailing silence, on word count, and whenever you send a `finalize` message. `finalize_on_words` and `max_words` are the word-count parameters. For how all finalization triggers fit together and the recommended setup per orchestration, see [Finalization and Endpointing](/models/speech-to-text/features/endpointing).

## Parameters

| Parameter           | Type                  | Default  | Description                                              |
| ------------------- | --------------------- | -------- | -------------------------------------------------------- |
| `finalize_on_words` | `"true"` \| `"false"` | `"true"` | Word-count-based finalization.                           |
| `max_words`         | integer               | Unset    | Word count at which the pending transcript is finalized. |

Both parameters are **WebSocket-only** - they don't apply to the pre-recorded HTTP endpoint, which always returns one final transcript per request.

## Related

* [Inverse Text Normalization (ITN)](/models/speech-to-text/features/inverse-text-normalization) - `finalize_on_words` is most useful in combination with ITN.
* [End-of-Utterance Timeout](/models/speech-to-text/features/end-of-utterance-timeout) - controls automatic finalization timing when `finalize_on_words` stays `true`.
* [Pulse STT WebSocket reference](/api-reference/models/speech-to-text/speech-to-text) - full parameter reference and the `finalize` / `close_stream` control messages.