# Your voice stays on your computer by default > By default Shuriken turns speech into text on your own computer with NVIDIA Parakeet, offline and private. Or use your own key for a live speech provider. Source: https://shuriken.sh/blog/choosing-speech-to-text By default Shuriken turns speech into text on your own computer with NVIDIA Parakeet, offline and private. Or use your own key for a live speech provider. [Story](https://shuriken.sh/blog) By Shuriken Updated 5 November 2026 2 min read **Contents** We talk to our agents a lot. "Make the empty state friendlier, and the button on the left should match the one in settings" is faster to say than to type, and it's easier to say it while looking at the screen than while looking at the keyboard. What you say while you work is often about client code, unreleased products and people. So we made a simple default: your voice is turned into text on your own computer. If you'd rather use a cloud model, that's your call, on your own key. ## How it works # Shuriken runs NVIDIA Parakeet TDT 0.6B v3, an open speech model, on your computer with the sherpa-onnx runtime. Press the mic, talk, and the words appear as you speak. With the local model the audio never goes to a speech service, to us, or to any other third party. - Offline. Once the model is on your computer it needs no connection. Trains, planes and bad hotel Wi-Fi are fine. - One download. The model comes down once, about 465 MB. You can see its size and remove it any time in Settings › Dictation. - No key, no bill. There's nothing to sign up for and nothing to pay per minute. Voice is in Free and Pro alike. - 25 languages. Parakeet v3 transcribes 25 European languages and works out which one you're speaking. ## Your phone and the browser # A phone or a browser can't run the model. When you dictate from the [phone web app](https://shuriken.sh/features/phone) or a browser, your voice travels end-to-end encrypted to your own computer, which turns it into text and sends the words back. When that goes through our relay, the relay sees only sealed bytes it can't read. (More on [what our cloud can and can't see](https://shuriken.sh/blog/security-what-the-cloud-sees).) In Shuriken you can dictate to any agent, turned into text on your own computer by default. [Try Shuriken](https://shuriken.sh/) ## Or bring your own provider # Some people already pay for a speech service, or want one of its languages. In Settings › Dictation you can pick ElevenLabs, AssemblyAI, Deepgram, Speechmatics or OpenAI instead, paste your own API key and press Test: Shuriken opens a short live session to check the provider takes it. - Live only. Every provider streams: your words appear while you speak. Providers that only transcribe a finished recording aren't offered. - Your account, your bill. Your computer streams the audio straight to that provider under your key. It never passes through a speech account of ours. - Silence is never sent. A voice detector on your computer holds back the quiet between sentences, so the provider hears only speech. - Your words. AssemblyAI, Deepgram, Speechmatics and OpenAI also get the words Shuriken has learned you use, so a project name stops coming out as something else. - Your key stays encrypted. It's kept in your encrypted vault (our cloud keeps a key that opens it) or your computer's keychain, and it's never written to a log. Switch back to Local model any time and nothing leaves your computer again. ## Paste while you talk # Our favourite trick works with every provider. Halfway through a sentence, paste an error message, a file path or a screenshot. The text lands exactly where you were in the sentence, not at the end, and images attach as you go. It's also what makes pointing at a page in the [built-in browser](https://shuriken.sh/features/browser) and saying the change feel natural. Everything else about voice is on the [Voice page](https://shuriken.sh/features/voice). To hear it for yourself, [try Shuriken](https://shuriken.sh/) and press the mic. Share [X](https://x.com/intent/post?url=https%3A%2F%2Fshuriken.sh%2Fblog%2Fchoosing-speech-to-text&text=Your%20voice%20stays%20on%20your%20computer%20by%20default)[LinkedIn](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fshuriken.sh%2Fblog%2Fchoosing-speech-to-text)[Hacker News](https://news.ycombinator.com/submitlink?u=https%3A%2F%2Fshuriken.sh%2Fblog%2Fchoosing-speech-to-text&t=Your%20voice%20stays%20on%20your%20computer%20by%20default)[Reddit](https://www.reddit.com/submit?url=https%3A%2F%2Fshuriken.sh%2Fblog%2Fchoosing-speech-to-text&title=Your%20voice%20stays%20on%20your%20computer%20by%20default)[E-mail](mailto:?subject=Your%20voice%20stays%20on%20your%20computer%20by%20default&body=https%3A%2F%2Fshuriken.sh%2Fblog%2Fchoosing-speech-to-text) ## Sources - [NVIDIA Parakeet TDT 0.6B v3 model card](https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3) (accessed 22 October 2026) - [sherpa-onnx, the speech runtime Shuriken runs it with](https://github.com/k2-fsa/sherpa-onnx) (accessed 22 October 2026) ## Frequently asked questions **Does my voice leave my computer?** Not by default. NVIDIA Parakeet turns it into text on your own computer, and with the local model no speech service or other third party hears it. If you pick a live speech provider and add your own key, your audio streams to that provider under your account instead. **Which speech providers can I use with my own key?** ElevenLabs, AssemblyAI, Deepgram, Speechmatics and OpenAI, all live: your words appear while you speak. Pick one in Settings › Dictation, paste your key and press Test. **Does dictation work offline?** Yes, with the local model, once it is on your computer. It downloads once, about 465 MB, and then needs no connection to turn speech into text. **What about my phone or a browser?** They have no model of their own. Your voice travels end-to-end encrypted to your own computer, which transcribes it, or streams it to the live provider you picked, and sends back the text. **Which languages does it understand?** Parakeet TDT 0.6B v3 transcribes 25 European languages, English, Dutch, German, French and Spanish among them, and works out which one you speak. The live providers cover their own lists of languages. **Do I need a key or a paid plan?** No. The local model needs no key and no account with anyone else, and voice is in Free and Pro alike. A key is only for a live provider you choose, and that provider bills you. Voice Talk to your agents. Privately. Dictation runs on your own computer by default, so no speech service hears you. Or add your own key for a live speech provider. [Try Shuriken](https://shuriken.sh/) [How it works](https://shuriken.sh/features/voice) ## Keep reading - [What's new in Shuriken 5.1: Update 27 Oct 2026 16 min read Shuriken 5.1 adds sign-in with Google, Apple or an email code, Allow and Always allow at a MacBook's notch, and billing and fuel in Sources.](https://shuriken.sh/blog/shuriken-5-1) - [Shuriken vs Emdash (2026): Comparison 27 Oct 2026 6 min read Emdash is a free, open-source agent workspace for Mac, Windows and Linux. Shuriken adds account switching, session limits, your phone and voice.](https://shuriken.sh/compare/emdash) - [Shuriken vs Paradev, formerly Superset (2026): Comparison 27 Oct 2026 6 min read Paradev (formerly Superset) is a source-available parallel-agent app for macOS. Shuriken also runs on Windows and Linux and switches accounts for you.](https://shuriken.sh/compare/superset) - [How MCP sign-ins travel between your devices: Story 2 Nov 2026 4 min read Sign in to an MCP server once and every device of yours has it, servers included. Our cloud stores it sealed, and keeps a key that can open it.](https://shuriken.sh/blog/mcp-sign-ins-sealed) - [How to review AI-generated code without reading every line: Story 27 Oct 2026 5 min read We review agent diffs tests first: would each test fail without the change? Then deletions, the diff's edges, secrets, and a run. Our real misses.](https://shuriken.sh/blog/review-ai-generated-code) - [Your agents run on your machines, not ours: Story 27 Oct 2026 3 min read Shuriken runs coding agents on your own computers and servers. Our cloud links devices and relays sealed messages. Why we chose local, and the trade-off.](https://shuriken.sh/blog/agents-on-your-machines) ## Free tools - [Claude limit reset tracker](https://shuriken.sh/tools/limit-resets/claude) - [Codex limit reset tracker](https://shuriken.sh/tools/limit-resets/codex)