Skip to content

Repository files navigation

Alpha Banner

Alpha: Offline Android Voice Recognition Module

A lightweight, production-ready sensory module for Android applications. This project provides a bridge between Google ML Kit Speech Recognition (Kotlin) and central application logic (Java).

Features

  • 100% Offline: Uses on-device ML models.
  • Modular Design: Isolated "Sensory Department" architecture.
  • Error Handling: 3-attempt recursive retry for silence/timeouts.
  • Java Bridge: Clean interface for Java-based projects like JARVIS.

How It Works

The project uses a Link class to act as a "Middle Manager." You simply pass a listener and wait for the text.

Example for Java

Link sensoryDept = new Link();

sensoryDept.startAlphaEars(text -> { System.out.println("Result: " + text); if (text.startsWith("ERROR")) { // Handle failure } else { // Process command } });

Example for Kotlin

val link = Link() link.startAlphaEars(object : Link.TaskListener { override fun onHeard(text: String) { println("User said: $text") } })

Credits & Author

  • Lead Developer: Santanu Sarkar
  • Project: Part of the JARVIS Ecosystem (Sensory Department).
  • AI Collaborator: Gemini (Google) - Assisted with bridge logic and error-handling optimization.

Special Thanks

  • Google ML Kit Team: For the Alpha Speech Recognition SDK.

  • Kotlin Coroutines Team: For background processing.

  • 🤝 How to Credit

This project is open-source under the MIT License. If you use this module in your own application, please include the following in your "About" or "Credits" section:

Voice Recognition powered by Alpha by Santanu Sarkar

About

An offline, modular voice recognition system for Android built with Google ML Kit.

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages