Open source on-device toolkit for speech-to-text, intent recognition and text-to-speech, aimed at real-time voice agents and apps.
DicioTeam/dicio-android
An open-source, on-device voice assistant for Android with speech and graphical replies and a set of built-in skills.
Open-source alternative to:Google Assistant
Siri
- Stars
- 1,521
- Forks
- 185
- Open issues
- 146
- Last commit
- 11 days ago
What is it?
Dicio is a free and open source voice assistant for Android. It interprets user speech and generates responses entirely on the device, so questions are not sent to a cloud service for processing. Answers can be spoken aloud and shown graphically on screen.
The assistant ships with a range of skills, including web search through DuckDuckGo, weather from OpenWeatherMap, lyrics, app launching, a calculator, contacts and calls, timers, navigation, translation through Lingva, jokes, media control, notifications and flashlight control. Speech recognition uses Vosk with small models of about 50 MB that download when needed, and wake word detection uses OpenWakeWord with a default "Hey Dicio" keyword. Users can import custom wake word models.
The app is available in more than a dozen languages, including English, German, French, Spanish, Russian and Chinese-unrelated European and Slavic languages such as Czech, Polish and Ukrainian. The project is distributed through F-Droid, GitHub releases and the Play Store.
Key features
- 01Runs speech recognition and command interpretation on the device.
- 02Provides both spoken and on-screen responses.
- 03Includes skills for search, weather, timers, calculator, navigation and translation.
- 04Supports custom wake word models in .tflite format.
- 05Available in about 14 languages, with speech models downloaded on demand.
Use cases
- Asking quick questions hands-free without sending voice data to a cloud assistant.
- Setting timers, checking the time or opening apps by voice on an Android phone.
- Using a custom wake word for a personalized voice assistant.
- Exploring or contributing to an open-source Android voice assistant codebase.
Quick start
- 1
Download the app from F-Droid, GitHub releases, or Play Store
- 2
Open the app and grant the microphone permission
Things to know
- Speech recognition models of about 50MB download automatically when needed.
- Some skills need external services such as OpenWeatherMap and Genius, which require their own access.
- Android only.
Topics
FAQ
- What is dicio-android?
- An open-source, on-device voice assistant for Android with speech and graphical replies and a set of built-in skills. Dicio is a free and open source voice assistant for Android. It interprets user speech and generates responses entirely on the device, so questions are not sent to a cloud service for processing. Answers can be spoken aloud and shown graphically on screen.
- Is dicio-android free to use? What license does it have?
- dicio-android is open source under the GPL-3.0 license. Check the license terms before using it in commercial products.
- Is dicio-android still maintained?
- Actively maintained. The last commit to dicio-android was on Oct 11, 2026.
- What can I use dicio-android for?
- Asking quick questions hands-free without sending voice data to a cloud assistant. Setting timers, checking the time or opening apps by voice on an Android phone. Using a custom wake word for a personalized voice assistant. Exploring or contributing to an open-source Android voice assistant codebase.
- How do I install dicio-android?
- Download the app from F-Droid, GitHub releases, or Play Store Open the app and grant the microphone permission
- Where can I get dicio-android?
- The source code and installation instructions are on GitHub: https://github.com/DicioTeam/dicio-android
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source voice dictation app for Mac: hold Fn, speak, and cleaned-up text is typed wherever your cursor is.
Native macOS speech-to-text app that transcribes dictation on-device or via cloud APIs, then reprocesses text with reusable AI workflows.
Leon is an open-source personal AI assistant that runs locally and can use tools, memory and context to complete tasks.
Cross-platform browser, Android and iOS client for the SEPIA personal voice assistant, with chat, speech input and TTS output.
Offline speech toolkit using onnxruntime for speech recognition, text-to-speech, VAD, diarization and more, across many platforms and languages.