wilbowes/EchoMuse

Alexa replacement and controller for Echo Dot 2nd Generation device.

What it solves

EchoMuse allows you to repurpose second-hand Amazon Echo Dot (2nd gen) hardware into a fully local, private voice assistant for Home Assistant. It removes the need for Amazon accounts, cloud services, and Alexa, turning the device into a local voice satellite that integrates directly with your existing Home Assistant Assist pipeline.

How it works

The project replaces the original Amazon software on the Echo Dot with either a custom Linux userspace (emOS) or a rooted version of the original FireOS. A controller component, which runs as a Home Assistant add-on or in a Docker container, manages the device, handles wake-word detection, and processes voice commands locally. The Dot communicates with your Home Assistant instance, allowing you to use local speech-to-text and text-to-speech engines like Whisper and Piper.

Who it’s for

This project is for privacy-conscious smart home enthusiasts who want to utilize inexpensive, older hardware to build a completely local, offline-capable voice control system within their Home Assistant ecosystem.

Highlights

  • Full Privacy: Operates entirely on your local network with no Amazon cloud dependency or telemetry.
  • Native Integration: Appears as a standard ESPHome voice satellite in Home Assistant.
  • Advanced Features: Supports wake-word barge-in, multi-device coordination (the closest device answers), and custom wake-word training.
  • Hardware Reuse: Extends the life of older smart speakers by providing a custom firmware and controller stack.
  • Flexible Setup: Offers a dedicated dashboard for managing device settings, EQ, LED ring behavior, and firmware updates.

Related

  • Project
    Softcatala/whisper-ctranslate2

    whisper‑ctranslate2 is a faster, CTranslate2‑backed command‑line client compatible with OpenAI’s Whisper. It offers up to 4‑16× speed‑ups, GPU/CPU support, live mic transcription, VAD filtering, speaker diarisation, and a ready‑to‑run Docker image.

  • Project
    mmorise/World

    A high-quality speech analysis, manipulation, and synthesis system that estimates pitch, aperiodicity, and spectral envelope to generate synthetic speech.

  • Project
    shivammehta25/Matcha-TTS

    Matcha-TTS is a fast, non-autoregressive text-to-speech architecture that uses conditional flow matching to enable high-quality, efficient speech synthesis.

  • Project
    AudileTeam/Audile

    Audile is an open-source Android app that identifies music playing nearby by integrating with multiple recognition services and providing direct links to streaming platforms.