Home Assistant Supertonic2 TTS Add-on

Ultra-fast, on-device multilingual text-to-speech powered by Supertonic2 for Home Assistant.

About

This repository contains a Home Assistant add-on that brings Supertonic2 TTS to your home automation setup. Supertonic2 is a lightning-fast, privacy-focused text-to-speech system that runs entirely on your device.

Why Supertonic2?

• ⚡ Blazingly Fast: Generates speech up to 167× faster than real-time
• 🔒 100% Private: All processing happens on your device - no cloud, no API calls, no data sent anywhere
• 🌍 Multilingual: Native support for English, French, Spanish, Portuguese, and Korean
• 🎤 Multiple Voices: Choose from 10 different voices (5 male, 5 female)
• 🎛️ Highly Configurable: Control speed, volume, quality, and voice selection
• 💪 Works Offline: Zero network dependency after initial setup

Features

Feature	Details
Languages	English (en), French (fr), Spanish (es), Portuguese (pt), Korean (ko)
Voices	M1, M2, M3, M4, M5 (male) • F1, F2, F3, F4, F5 (female)
Speed Control	0.5× to 2.0× (50% slower to 2× faster)
Volume Control	1.0× to 3.0× amplification
Quality Levels	1-10 (higher = better quality, slightly slower)
Audio Format	WAV, 44.1kHz, mono
Protocol	Wyoming protocol with automatic discovery

Installation

Method 1: Add Repository to Home Assistant (Recommended)

Quick Add (One Click):

Or manually:

1. In Home Assistant, navigate to Settings → Add-ons → Add-on Store
2. Click the ⋮ menu (three dots) in the top right corner
3. Select Repositories
4. Add this repository URL:

   https://github.com/JulienDeveaux/HA-Supertonic-Addon
   

5. Find "Supertonic2 TTS" in the add-on list
6. Click Install

Method 2: Manual Installation (Development/Testing)

1. Clone this repository:

   git clone https://github.com/JulienDeveaux/HA-Supertonic-Addon.git
   cd HA-Supertonic-Addon

2. Copy the supertonic_tts directory to your Home Assistant addons folder:

   # For Home Assistant OS:
   cp -r supertonic_tts /addons/
   
   # For Supervised:
   cp -r supertonic_tts /usr/share/hassio/addons/local/

3. Restart the Supervisor:

   ha supervisor reload

4. The add-on will appear in your add-on store

Quick Start

1. Configure the Add-on

After installation, configure your preferences:

default_language: "fr"      # en, fr, es, pt, ko
default_voice: "M4"         # M1-M5, F1-F5
speed: 1.5                  # 0.5 - 2.0
volume_boost: 2.0           # 1.0 - 3.0
quality: 5                  # 1 - 10

2. Start the Add-on

• Click Start
• Enable Start on boot (optional)
• Enable Watchdog (optional)

⏱️ Note: First start takes 2-5 minutes to download models (~250MB)

3. Automatic Discovery

That's it! Home Assistant will automatically discover the TTS service via Wyoming protocol.

• The integration appears automatically in Settings → Devices & Services
• Click Configure when you see the Wyoming Protocol notification
• All 50 voices (5 languages × 10 voices) will be available

Architecture Support

Architecture	Supported	Notes
aarch64	✅ Yes	ARM 64-bit (Raspberry Pi 3/4/5, etc.)
amd64	✅ Yes	Intel/AMD 64-bit (most PCs)
armv7	✅ Yes	ARM 32-bit v7
armhf	❌ No	Use armv7 instead
i386	❌ No	32-bit x86 not supported

Wyoming Protocol

This add-on uses the Wyoming protocol for communication with Home Assistant:

• Port: 10300 (TCP)
• Protocol: Wyoming (binary protocol for voice services)
• Discovery: Automatic via Zeroconf/mDNS (_wyoming._tcp.local.)
• Communication: Event-based (Describe, Synthesize, AudioChunk, AudioStop)

No REST API or HTTP endpoints are exposed. All communication happens through the Wyoming protocol.

Performance Benchmarks

Generation speed examples (Apple M4 Pro):

Quality	Real-Time Factor	Use Case
2	167×	Quick notifications
5	51×	General use (recommended)
10	17×	High-quality audio

Real-Time Factor (RTF): 1× means generation takes the same time as playback. Higher is faster.

System Requirements

Requirement	Minimum	Recommended
RAM	512MB	1GB+
Disk Space	500MB	1GB+
CPU	ARM/x86 dual-core	Quad-core+
Network	Required for setup only	-

Supported Languages

Language	Code	Native Accent	Status
English	en	✅ Yes	Fully supported
French	fr	✅ Yes	Fully supported
Spanish	es	✅ Yes	Fully supported
Portuguese	pt	✅ Yes	Fully supported
Korean	ko	✅ Yes	Fully supported

Contributing

Contributions are welcome! Please:

1. Fork this repository
2. Create a feature branch
3. Make your changes
4. Submit a pull request

For bugs or feature requests, please open an issue.

Credits

• Supertonic2 - Created by Supertone Inc.
  • GitHub: supertone-inc/supertonic
  • HuggingFace: Supertone/supertonic-2