Compare
The programs people compare us with most often, in one table. Honest, including where we lose. Prices and features are what each maker publishes.
| HoldToType | Handy | Wispr Flow | Dragon | Whisper Flow | DictaFlow | Typeless | |
|---|---|---|---|---|---|---|---|
| Code-signed builds | Not yet i | Yes | Yes | Yes | Yes | Yes | Yes |
| Price | Free | Free | $12 / month | $699 once | From $9 / month | $7 / month | $12 / month |
| Subscription or credits | None | None | Subscription | One-time licence | Credits | Subscription | Subscription |
| Account | None | None | Required | Required | Required | Required | Required |
| Audio leaves your computer | No | No | Yes | No | Yes | Yes | Yes |
| Works offline | Yes | Yes | No | Yes | No | Paid option | No |
| Open source | MIT | MIT | No | No | No | No | No |
| Languages | 99 | 99 | 100+ | 7 | 100+ | 90+ | 100+ |
| Live text while speaking | Yes | No | No | Yes | No | No | Yes |
| Translation | 7 languages | No | No | No | No | No | Yes |
| Editing by a local model | Yes | No | Cloud | No | Cloud | Cloud | Cloud |
| Choice of speech model | 9 models, one per language | Whisper sizes and Parakeet | No, theirs | No, theirs | No, theirs | No, theirs | No, theirs |
| Your own speech models | Any Whisper GGML file or sherpa-onnx model put in the models folder | — | No | No | No | No | No |
| Models that show text while you speak | Nemotron 3.5 | No | No | Its own engine | No | No | Its own engine |
| Post-processing by a model | Yes | No | Yes | No | Yes | Yes | Yes |
| Where post-processing runs | On your disk, or a server you choose | — | Their cloud | — | Their cloud | Their cloud | Their cloud |
| Choice of editor model | Any GGUF from Hugging Face, or any API | No | No | No | No | No | No |
| Your own prompts | Yes, chains, each with a shortcut | No | Built-in styles | No | — | — | Built-in edits |
| Recognition server on another computer | Yes | No | No | No | No | No | No |
| Voice commands you define | Yes | No | Built-in only | Yes | — | — | Built-in only |
| Where dictation history is kept | On your disk, off by default | On your disk | Their servers | — | Their servers | Their servers | Their servers, on-device option |
| Interface languages | 8 | — | English | Several | — | — | — |
| Custom vocabulary | Yes | No | Yes | Yes | Yes | Yes | Yes |
| Portable folder | Yes | No | No | No | No | No | No |
Prices and features as published by each maker in September 2026. A dash means the maker does not say.
No Mac and no phone. The builds are not code-signed yet, so Windows warns once. There is no cloud sync between your computers: settings and history live in the folder, and you copy them yourself.
Accuracy on everyday speech is close across the board: the same open models sit under most of these programs. Custom vocabulary is common. The paid ones ship a polished onboarding that a one-person project cannot match.
Nothing to pay and nothing to sign up for. The audio never leaves the computer. The source is open under MIT. A model per language, live text while you speak, translation into seven languages, an editor model that runs on your disk, and a portable folder you can carry on a stick.
Windows voice typing, the built-in Win+H tool, is free and needs no install, but for most languages it sends the audio to Microsoft and it has no dictionary of your own words. Mac-only programs such as VoiceInk, FluidVoice or HoldType, the one with the similar name, are left out because HoldToType is for Windows. Vibe and other file transcribers do a different job: they transcribe recordings, not what you say at the cursor.