Local dictation, end to end.
Whisperstream is a complete dictation toolkit for Windows. Every feature below runs on your own PC, with no cloud and no account. Core transcription is on-device by default; an optional AI cleanup step is the only piece that can reach out, and only if you turn it on.
The everyday pipeline: capture your voice, transcribe it on your CPU, and type it into whatever app you have open.
Whisperstream transcribes your speech entirely on your computer using NVIDIA's open-weight Parakeet TDT v3 model running on your CPU. Your audio is never uploaded, streamed, or stored on a server, and you do not need an account. After the one-time model download, transcription works with no internet connection at all.
Dictation is push-to-talk by default: hold your hotkey, speak, and release, so nothing is ever listening in the background. The default is Right Shift, and you can remap it to any key from the Controls tab. Prefer hands-free? Switch to toggle mode, where one press starts recording and another stops it. No admin rights are needed to register the shortcut.
Whisperstream pastes your transcribed text into whatever window has focus, so it works the same in Outlook, Word, Slack, VS Code, your browser, or a terminal. There is no add-in to install and no per-app configuration. If your cursor can blink in a text field, Whisperstream can type into it.
Whisperstream keeps a private history of everything you dictate, so you can find a past transcript, replay its original audio, and copy the result again later. The history is encrypted at rest on your own device and is never synced to a server. Search by keyword, play the recording back next to its text, and delete anything you would rather not keep.
Beyond live dictation, Whisperstream transcribes audio files you already have. Drop in a meeting recording, a voice memo, or an interview and it runs through the same on-device Parakeet model, with no upload and no account. The finished transcript lands in your history alongside everything you have dictated, ready to search, replay, and copy.
Whisperstream transcribes 25 languages, including English, Spanish, French, German, Italian, Portuguese, Russian, and Polish, plus 17 more. Every language runs through the same on-device model, so multilingual dictation stays fully local with no cloud service and no per-language download.
Whisperstream needs the internet only twice: once to download the speech model and once to activate your license. After that, dictation, transcription, and formatting all run with no connection at all. You can dictate on a plane, in a secure facility, or anywhere your network is down.
Whisperstream can lower or mute your system volume while you dictate, so music or a video does not bleed into your recording or distract you. Choose off, reduce to 20 percent, or full mute in the Audio settings. Volume returns to its previous level the moment you release the hotkey.
Optional tools that clean, shape, and tune your output. All off by default, all yours to switch on.
Whisperstream can optionally clean up your dictation: it removes filler words, fixes grammar and punctuation, and applies your spoken self-corrections (say "Friday, sorry Saturday" and it keeps just "Saturday"). The cleanup model runs locally on a capable GPU, with a free cloud fallback for older hardware. It is off by default, so you choose if and when to turn it on.
Each profile is its own AI prompt, and Whisperstream switches between them automatically by detecting the focused window. Start from the built-in modes or write your own profile with any prompt you want: terse for your terminal, polished for email, casual for chat. Route by app or by a specific website, and it falls back to your default everywhere else.
The dictionary lets you add word-for-word overrides so Whisperstream spells names, technical terms, and acronyms the way you want them. Add an entry once in the Dictionary tab and it applies to every transcription after. Overrides are case-insensitive and matched as you dictate, so you stop fixing the same word by hand.
Beyond the built-in modes, you can give the AI cleanup your own prompt with custom formatting or style instructions, and every enhanced transcription comes back shaped to match. Keep a formal tone, stay terse, use British spelling, format as bullet points, or follow your own house style. Your prompt runs on a local model or your own cloud key, automatically, so dictated text lands already styled the way you want.
What Whisperstream runs on, and why the install is one you can trust.
The Whisperstream installer is code-signed with a Microsoft-trusted certificate, so Windows recognizes a verified publisher instead of an unknown one. You get a clean, trusted install, and the signature confirms the download came from us and was not tampered with.
Core transcription runs entirely on your CPU, so you do not need a dedicated graphics card. A modern processor delivers near-instant results; a GPU is only used for the optional AI cleanup feature. The app works on 8 GB of RAM, with 16 GB recommended.
Free to try. No account.