Supertonic TTS Flutter

Multilingual text-to-speech (TTS) for Flutter. Powered by ONNX Runtime for fast local inference.

Note: This is an unofficial Flutter port of the Supertonic project.

Features

• 🌍 Multilingual Support - English, Korean, Spanish, Portuguese, and French
• 🎭 Multiple Voice Styles - 10 voices (5 male, 5 female)
• ⚡ Local Processing - Runs fully on-device
• 🎛️ Customizable - Adjustable speech speed and quality settings
• 📱 Cross-Platform - Supports Android, iOS, macOS, Linux, and Web
• 🔊 Neural TTS - Based on diffusion models

Platform Support

Platform	Status	Minimum Version
Android	✅ Supported	API 21+ (Android 5.0)
iOS	✅ Supported	iOS 16.0+
macOS	✅ Supported	macOS 14.0+
Linux	✅ Supported	Ubuntu 20.04+
Web	✅ Supported	Modern browsers

Web Setup

For Web builds, flutter_onnxruntime requires ONNX Runtime Web to be loaded in your app's web/index.html before flutter_bootstrap.js:

<script src="https://cdn.jsdelivr.net/npm/onnxruntime-web@1.22.0/dist/ort.min.js"></script>
<script src="flutter_bootstrap.js" async></script>

Installation

Add this to pubspec.yaml:

dependencies:
  supertonic_flutter: ^1.0.0

Quick Start

import 'package:supertonic_flutter/supertonic_flutter.dart';

final tts = SupertonicTTS();
await tts.initialize();

final result = await tts.synthesize(
  'Hello, world!',
  language: 'en',
  voiceStyle: 'M1',
);

initialize() uses this order:

• cached models
• bundled assets
• download from Hugging Face if needed

Model Setup

Option A: Auto-Download (Recommended)

No manual setup required. Models are downloaded from Hugging Face on first use (~268 MB).

// Just initialize — models download automatically if not found
final tts = SupertonicTTS();
await tts.initialize();

To pre-download at a convenient time:

if (!await SupertonicTTS.modelsReady()) {
  await SupertonicTTS.preDownloadModels(
    onProgress: (done, total, file, progress) {
      print('[$done/$total] $file: ${(progress * 100).toInt()}%');
    },
  );
}

Option B: Bundle Assets Manually

Download the model files from Hugging Face and add them to your app's assets. This avoids runtime downloads but increases app size by ~268 MB.

assets/
├── onnx/
│   ├── duration_predictor.onnx
│   ├── text_encoder.onnx
│   ├── vector_estimator.onnx
│   ├── vocoder.onnx
│   ├── tts.json
│   └── unicode_indexer.json
└── voice_styles/
    ├── M1.json, M2.json, M3.json, M4.json, M5.json
    └── F1.json, F2.json, F3.json, F4.json, F5.json

Update pubspec.yaml:

flutter:
  assets:
    - assets/onnx/
    - assets/voice_styles/

Platform Configuration

This package uses flutter_onnxruntime, which needs a few platform-specific settings:

➡️ Android

Add a ProGuard rule so ONNX Runtime classes are not obfuscated.

Create or edit android/app/proguard-rules.pro:

-keep class ai.onnxruntime.** { *; }

Or run this command from your terminal:

echo "-keep class ai.onnxruntime.** { *; }" > android/app/proguard-rules.pro

See the flutter_onnxruntime troubleshooting guide for more details.

➡️ iOS

ONNX Runtime requires iOS 16+ and static linkage.

In ios/Podfile, update the following lines:

platform :ios, '16.0'

# ... existing code ...

target 'Runner' do
  use_frameworks! :linkage => :static

  # ... existing code ...
end

And add to the post_install hook:

post_install do |installer|
  installer.pods_project.targets.each do |target|
    flutter_additional_ios_build_settings(target)
    target.build_configurations.each do |config|
      config.build_settings['IPHONEOS_DEPLOYMENT_TARGET'] = '16.0'
    end
  end
end

➡️ macOS

macOS builds require macOS 14+.

In macos/Podfile, change:

platform :osx, '14.0'

Update the "Minimum Deployments" to 14.0 in Xcode:

open macos/Runner.xcworkspace

Then in Xcode:

1. Select Runner project in the navigator
2. Select Runner target
3. Go to General tab
4. Change Minimum Deployments to 14.0

Also add to the post_install hook in your Podfile:

post_install do |installer|
  installer.pods_project.targets.each do |target|
    flutter_additional_macos_build_settings(target)
    target.build_configurations.each do |config|
      config.build_settings['MACOSX_DEPLOYMENT_TARGET'] = '14.0'
      config.build_settings['ALLOW_STATIC_FRAMEWORK_TRANSITIVE_DEPENDENCIES'] = 'YES'
    end
  end
end

Demo

See example/ for a demo app with language selection, voice switching, and adjustable settings.

More Examples

Basic Example

import 'package:supertonic_flutter/supertonic_flutter.dart';

// Initialize
final tts = SupertonicTTS();
await tts.initialize();

// Synthesize speech
final result = await tts.synthesize(
  'Hello, world!',
  language: 'en',
  voiceStyle: 'M1',
);

// Convert to WAV and save
final wavBytes = result.toWavBytes();
final file = File('output.wav');
await file.writeAsBytes(wavBytes);

With Audio Player

import 'package:supertonic_flutter/supertonic_flutter.dart';

final tts = SupertonicTTS();
final player = TTSAudioPlayer();

// Initialize
await tts.initialize();

// Synthesize and play
final result = await tts.synthesize(
  'Hello, this is a test.',
  language: 'en',
  voiceStyle: 'F1',
  config: TTSConfig(
    speechSpeed: 1.05,
    denoisingSteps: 5,
  ),
);

await player.play(result);