SpeechDataBuilder

KO | EN

gitlite — search

#javascript #typescript #go #golang #android #deep-learning #linux #cli #kubernetes #vue #awesome-list #pytorch #security #open-source

SpeechDataBuilder

(★ 10)

Browser-based open-source tool for creating high-quality TTS/STT datasets. Features AI transcription, multiple export formats, and audio visualization with no server-side processing. Perfect for ML engineers, developers, and researchers.

SpeechDataBuilder Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Speech Data Builder: TTS & STT Dataset Creation Tool

![Speech Data Builder Logo](https://raw.githubusercontent.com/FS-17/SpeechDataBuilder/main/public/assets/logo_name.png)

## The Ultimate Free Tool for Creating High-Quality Speech Datasets

Speech Data Builder is a powerful, open-source web application designed to streamline the creation of professional speech datasets for Text-to-Speech (TTS) and Speech-to-Text (STT) model training. Whether you're developing voice assistants, speech recognition systems, or voice synthesis applications, this tool provides everything you need to build high-quality speech corpora.

**Live demo:** [https://fs-17.github.io/SpeechDataBuilder/](https://fs-17.github.io/SpeechDataBuilder/)

## 🔥 Key Features

- **AI-Powered Transcription**: Automatically transcribe your audio files using state-of-the-art AI models from Google AI Studio and OpenAI.
- **Multiple Export Formats**: Export your datasets in popular formats including LJSpeech, Common Voice, and custom configurations.
- **Audio Visualization**: Precise waveform display with region selection for accurate transcript alignment.
- **Text Normalization**: Automatic text normalization for TTS model training with special handling for multiple languages.
- **Batch Processing**: Efficiently process multiple audio files in one go, saving time and effort.
- **Real-time Editing**: Edit transcripts with auto-save and instant audio playback.
- **Import Existing Metadata**: Import LJSpeech, CSV, TSV, JSON, or TXT transcripts and match them to uploaded audio.
- **Customizable Settings**: Configure formats, AI providers, and normalization to match your dataset requirements.
- **Dark Mode Support**: Work comfortably in any lighting conditions with dark mode.
- **PWA / Offline**: Installable app with a cached shell for offline use after first load.

## 📊 Perfect For

- **ML Engineers**: Building custom voice models and speech recognition systems
- **Voice Assistant Developers**: Creating training data for voice assistants
- **Researchers**: Collecting and organizing speech data for linguistic research
- **Content Creators**: Transcribing podcasts, interviews, or educational content
- **Language Preservationists**: Documenting endangered languages and dialects

## 📸 Screenshots

### Light Mode

![SpeechDataBuilder Light Mode - TTS Dataset Creation](https://raw.githubusercontent.com/FS-17/SpeechDataBuilder/main/public/assets/screenshot-light.jpg)

### Dark Mode

![SpeechDataBuilder Dark Mode - Speech Recognition Dataset Tool](https://raw.githubusercontent.com/FS-17/SpeechDataBuilder/main/public/assets/screenshot-dark.jpg)

## 🚀 Getting Started

1. Open the application in your web browser
2. Upload your audio files (MP3, WAV, OGG, FLAC supported)
3. Use AI transcription or manually enter transcripts
4. Customize and normalize text as needed
5. Export in your preferred format

### Option 1: Use the Live Version

Visit [https://fs-17.github.io/SpeechDataBuilder](https://fs-17.github.io/SpeechDataBuilder) to start creating speech datasets immediately.

### Option 2: Run Locally

```bash
git clone https://github.com/FS-17/SpeechDataBuilder.git
cd SpeechDataBuilder
npm install
npm run dev
```

Open the URL Vite prints (usually `http://localhost:5173/`).

| Command           | Description                  |
| ----------------- | ---------------------------- |
| `npm run dev`     | Start the Vite dev server    |
| `npm run build`   | Typecheck + production build |
| `npm run preview` | Preview the production build |
| `npm run lint`    | Run oxlint                   |

### Production build (GitHub Pages)

```bash
npm run build
npm run preview
```

Build output is in `dist/`. The production `base` path is `/SpeechDataBuilder/` for deployment at  
https://fs-17.github.io/SpeechDataBuilder/

## 💡 Advanced Features

### LJSpeech Dataset Creation

Speech Data Builder excels at creating datasets in the popular LJSpeech format, widely used for training TTS models. The tool automatically:

- Normalizes text (converting numbers to words, removing special characters)
- Handles non-Latin scripts with specialized normalization
- Generates properly formatted metadata files

### AI Integration

Connect with leading AI services to automate the transcription process:

- **Google AI Studio**: Leverage Gemini models for accurate transcription and text normalization
- **OpenAI**: Utilize Whisper and GPT models for speech recognition and text normalization

## 📝 Speech Dataset Creation Guide

### Step 1: Add Audio Files

- Click the "Upload Audio Files" button or drag and drop audio files onto the upload area
- Supported formats: MP3, WAV, OGG, FLAC
- Organize your speech samples for efficient processing

### Step 2: Transcribe Audio for Speech Recognition

- Select a file from the list to load it in the editor
- Use the audio playback controls to listen to the speech
- Type the accurate transcript in the transcript editor
- For TTS datasets, ensure proper punctuation and formatting
- Click "Save" or use Ctrl+S to save
- Optionally use **AI Transcribe** after configuring a provider in Settings

### Step 3: Export Speech Dataset

- Navigate to the **Export** tab
- Choose your preferred file format (CSV, JSON, TXT)
- Transcript layout follows the format chosen in Settings (LJSpeech, Common Voice, default, or custom)
- For TTS datasets: LJSpeech includes normalized text
- For STT datasets: CSV or JSON provide flexible options
- Optionally include audio files in the export (ZIP)
- Click **Export Dataset** to download your speech data

## ⌨️ Keyboard Shortcuts for Fast Dataset Creation

| Shortcut   | Action                            |
| ---------- | --------------------------------- |
| Space      | Play/Pause audio                  |
| ← / →      | Skip back/forward 1 second        |
| A / D      | Skip back/forward 5 seconds       |
| W / S      | Increase/decrease playback speed  |
| R          | Create region at current position |
| Ctrl+S     | Save transcript                   |
| Ctrl+Alt+T | Toggle dark/light theme           |

## 📋 Speech Dataset Export Formats

### LJSpeech Format (Popular for TTS)

```
filename|transcript text|normalized text
```

### CSV Format (Flexible for STT)

```
FileName,Transcript
file1.wav,"This is the transcript for file one."
```

Also supports Common Voice–style CSV and custom delimiter/template formats in Settings.

## 🤖 AI-Assisted Speech Transcription

Speech Data Builder supports AI-powered transcription to speed up dataset creation:

1. Navigate to the **Settings** tab
2. Select your preferred AI service provider
3. Enter your API key and choose a model
4. Return to the workspace and click **AI Transcribe**
5. Edit the AI-generated transcript for perfect accuracy
6. Create large speech datasets in less time

API keys stay in `localStorage` on this device only.

## 🌐 Browser Compatibility

Speech Data Builder works in all modern browsers for cross-platform speech dataset creation:

- Chrome (recommended)
- Firefox
- Edge
- Safari

## 👥 Contributing to Speech Data Builder

Contributions are welcome! Help improve this open-source speech dataset tool:

1. Fork the repository
2. Create your feature branch (`git checkout -b feature/amazing-feature`)
3. Commit your changes (`git commit -m 'Add some amazing feature'`)
4. Push to the branch (`git push origin feature/amazing-feature`)
5. Open a Pull Request

## 📝 License

This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.

## 🙏 Acknowledgments

- [WaveSurfer.js](https://wavesurfer-js.org/) for audio visualization
- [Bootstrap](https://getbootstrap.com/) for UI components
- [Font Awesome](https://fontawesome.com/) for icons
- [JSZip](https://stuk.github.io/jszip/) for file export capabilities
- [Vite](https://vite.dev/) / [React](https://react.dev/) / [Zustand](https://zustand-demo.pmnd.rs/)

## 🔧 Technical Details

Speech Data Builder runs entirely in your browser with no server-side processing of your audio files. Transcripts and settings are stored in **localStorage** on your device.

**Stack:**

- React 19 + TypeScript + Vite
- Zustand for state
- WaveSurfer.js for waveform / regions / trim
- Bootstrap 5 for responsive UI
- vite-plugin-pwa for installable offline-capable shell
- JSZip (lazy-loaded) when exporting a ZIP with audio

**Privacy note:** Audio and transcripts stay in the browser; AI transcription (when enabled) sends audio to the provider you configure (Google AI Studio or OpenAI). The hosted demo may load Google Analytics for site usage metrics.

## 📞 Contact

For questions, feedback, or support, please [open an issue](https://github.com/FS-17/SpeechDataBuilder/issues) on our GitHub repository.

---

**Speech Data Builder** — Making voice dataset creation accessible to everyone.

_Keywords: TTS dataset, STT dataset, speech recognition data, voice dataset creator, LJSpeech format, speech corpus, AI voice training_

## What’s new

### v2.0 (React rewrite)

- Full rewrite to **React 19 + TypeScript + Vite**
- Zustand stores for files, transcripts, settings, and UI
- Import workflow: LJSpeech, CSV, TSV, JSON, TXT
- Waveform regions, trim-to-selection, restore original
- AI model picker for Google Gemini and OpenAI Whisper
- PWA install + offline app shell
- Keyboard shortcuts and accessibility improvements (skip link, focus outlines)

### Earlier highlights

- Offline support (PWA): installable app with cached core assets
- Keyboard shortcuts: Space, ←/→, A/D, W/S, R, Ctrl+S, Ctrl+Alt+T
- Accessibility: skip link to main content, improved focus outlines
- SEO: canonical link and social sharing image

### Install as an app (optional)

Open the site in Chrome/Edge and use “Install app” from the address bar. Works offline after first load.

SpeechDataBuilder

SpeechDataBuilder Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

SpeechDataBuilder

SpeechDataBuilder Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README