# Wispr flow / third part speech to text API

**URL:** https://forum.rabbitcommunity.tech/t/wispr-flow-third-part-speech-to-text-api/15870
**Category:** general chat
**Created:** [July 24, 2026, 7:16am UTC](https://forum.rabbitcommunity.tech/t/wispr-flow-third-part-speech-to-text-api/15870 "2026-07-24T07:16:09Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![gael](https://avatars.discourse-cdn.com/v4/letter/g/a698b9/32.png) [@gael](https://forum.rabbitcommunity.tech/u/gael)
#### Post date: [July 24, 2026, 7:16am UTC](https://forum.rabbitcommunity.tech/t/wispr-flow-third-part-speech-to-text-api/15870/1 "2026-07-24T07:16:09Z")

</div>

Transcription accuracy on the Rabbit R1 is often inconsistent. As a long-time user of wispr flow, I would appreciate the ability to select alternative models for processing audio transcription via Bring Your Own Key (BYOK).