Building an Open Source Android Voice Assistant with Kotlin

Source: Dev.to

Replace Google Assistant with Your Own AI

What if you could long‑press your Home button and talk to YOUR AI instead of Google’s? I built OpenClaw Assistant – an open‑source Android app that does exactly that.

• 📹 Demo:
• 🔗 GitHub:

Features

• 🏠 System Assistant Integration – long‑press Home to activate
• 🎤 Custom Wake Words – “Jarvis”, “Computer”, or any phrase you choose
• 📴 Offline Wake Word Detection – powered by Vosk, no cloud required
• 🔊 Voice I/O – speech recognition + TTS
• 🔗 Backend‑agnostic – connect to Ollama, OpenAI, Claude, or any custom API

Architecture Overview

• The app registers as Android’s digital assistant.
• Vosk listens for wake words locally.
• On activation, speech is transcribed and sent to your webhook.
• The response is spoken via TTS.

Installation

git clone https://github.com/yuga-hashimoto/OpenClawAssistant
cd OpenClawAssistant
./gradlew assembleDebug

Or download the APK from the Releases page on GitHub.

Webhook API Contract

Your backend must accept a POST request:

POST /your-endpoint
Content-Type: application/json

{
  "message": "user's speech",
  "session_id": "..."
}

and respond with:

{
  "response": "AI's reply"
}

Contributing

Contributions are welcome! Feel free to open issues or submit pull requests on the GitHub repository.