Free, open-source Android keyboard for voice dictation — 100% on-device.
Speak in any app. No cloud, no account, no subscription.
-3DDC84?logo=android&logoColor=white){kind=icon}
Website · Download APK · iOS · Desktop · Community
Dictus is a free, open-source Android keyboard that adds voice dictation to any app. All speech recognition runs on-device via Whisper (whisper.cpp) and NVIDIA Parakeet (sherpa-onnx) — no server, no account, no subscription.
- 🔒 100% on-device — your voice never leaves your phone. No cloud, no telemetry, no account.
- 🆓 Free & open source — MIT licensed, no subscription, fully auditable code.
- ⌨️ System-wide IME — works in every app as your default keyboard.
- ⚡ Multi-engine — Whisper (multilingual) or Parakeet (English, fast).
- 🌐 FR + EN dictionaries — smart word predictions while typing.
| Feature | Dictus | Wispr Flow | Gboard Voice | SuperWhisper |
|---|---|---|---|---|
| Price | Free | Free / $15/mo | Free | Free / $8.49/mo |
| 100% offline | ✅ | ❌ | ||
| Privacy-first | ✅ | ❌ | ||
| Open source | ✅ | ❌ | ❌ | ❌ |
| System keyboard | ✅ | ❌ | ✅ | ❌ |
| Cross-platform | ✅ (iOS · Android · Desktop) | iOS · macOS · Win · Android | Android · Wear OS | iOS · macOS · Win |
See the full comparison on getdictus.com.
Dictus is currently in public beta — install by sideloading the APK from GitHub Releases.
- On your Android device, go to Settings → Apps → Special app access → Install unknown apps and allow your browser.
- Download the latest APK from Releases.
- Open the
.apkand tap Install. - Go to Settings → System → Languages & input → On-screen keyboard → Manage on-screen keyboards.
- Enable Dictus.
- Open any text field, tap the keyboard icon in the navigation bar, and select Dictus.
Requirements: Android 10 (API 29) or higher · ~150 MB for the smallest Whisper model.
| Keyboard | Model Manager | Settings |
|---|---|---|
 |
 |
 |
- Offline voice dictation — Whisper + NVIDIA Parakeet, entirely on-device
- Multi-engine STT — Whisper (multilingual) or Parakeet (English, fast)
- Smart suggestions — word predictions from FR+EN dictionaries while typing
- Personal dictionary — learns your frequently typed words
- System keyboard — works in any app as your default IME
- AZERTY & QWERTY — switchable keyboard layouts
- On-device Whisper + Parakeet engines
- System IME with AZERTY / QWERTY layouts
- FR + EN word predictions and personal dictionary
- Smart Mode Pro — on-device LLM reformulation
- Custom vocabulary (technical terms, names)
- Searchable local transcription history
- Audio-file transcription
- Sync settings across Dictus iOS / Android / Desktop (offline-first)
Have an idea? Open a feature request — we prioritize the most-upvoted ones.
- Kotlin + Jetpack Compose
- whisper.cpp (MIT) — Whisper STT
- sherpa-onnx (Apache 2.0) — Parakeet STT
- Material Design 3
Contributions are welcome — see CONTRIBUTING.md for build setup, module overview, and PR guidelines. Good entry points:
good first issueandhelp wantedin Issues- Bug reports with logs from a recent build
- Translations & locale tuning
Dictus collects no user data. All speech processing happens on your device. See our Privacy Policy.
Dictus is free and will stay free. If it helps you every day, consider supporting development — it directly funds new features and platform support.
MIT — see LICENSE.
Made with ❤️ by PIVI Solutions · @getdictus