# Someone did their own Otter.ai: Handy, an offline speech-to-text app

> Source: <https://dev.to/sam_notaltman/someone-did-their-own-otterai-handy-an-offline-speech-to-text-app-53lc>
> Published: 2026-11-19 15:10:13+00:00

**Key takeaway:** Handy shows a fully offline speech-to-text desktop app can be shipped in Rust across Windows, macOS and Linux, letting users keep dictation local instead of sending audio to a hosted service.

I run [I Did My Own](https://idmo.app/?utm_source=devto&utm_medium=social&utm_campaign=handy-spotlight), a directory of open-source projects by people who looked at a well-known product and built their own. Now and then I'll pick one project from it and write a short note about it. This one isn't mine: Handy is by [cjpais](https://github.com/cjpais/Handy).

[Handy](https://idmo.app/p/handy?utm_source=devto&utm_medium=social&utm_campaign=handy-spotlight) is a free, extensible, fully offline speech-to-text desktop app. On idmo it's listed as someone's own take on Otter.ai, Wispr Flow and Superwhisper.

From its idmo page today:

Hosted transcription and dictation tools are polished and convenient, and they're the reason a lot of us got used to talking to our computers. Handy takes a different angle: it's built to be fully offline, so transcription runs on your own machine instead of a hosted service. For people who dictate notes, code comments or messages all day and would rather keep that local, that's a meaningful trade-off.

It's also a nice example of a desktop app written in Rust that ships for all three major operating systems.

Do you use speech-to-text in your dev workflow? If you've tried an offline setup, what are you running, and what would you want from it?

And if you've built your own version of a well-known app, you can add it at [idmo.app/submit](https://idmo.app/submit?utm_source=devto&utm_medium=social&utm_campaign=handy-spotlight).
