Voice Arena allows you to generate TTS (Text-to-Speech) using two AI voice models, either randomly or by selection, vote for the better model, generate personal rankings based on voting results, and provides features like voice-to-voice conversion and voice cloning to create your own customized voice models.

中文 | English | 日本語

This is the open-source version of the Voice Arena from 302.AI. You can directly log in to 302.AI to use the online version with zero code and zero background, or modify and deploy it yourself according to your requirements.

In voice battles, you can randomly select or choose two voice models for text-to-speech generation, vote for the better model, and model names in random mode will be revealed after voting.

Select voice models and set parameters, AI will generate speech based on input text, supporting batch generation.

Select voice models and set parameters, AI will convert uploaded audio files or recordings into speech with the corresponding voice.

Use customized voice models for text-to-speech generation to expand your voice library.

Create custom voice models by uploading audio files or recordings.

The system automatically generates personal model rankings based on user voting results.

All generated speech can be found in the history records.

Supports various TTS models, including but not limited to OpenAI, Azure, Doubao, etc.

• Single text, single voice
• Single text, multiple voices
• Multiple texts, single voice
• Multiple texts, multiple voices

Supports voice-to-voice conversion, converting any voice to target voice.

Quickly clone new voice models through voice samples.

Generate voice model rankings based on user votes.

• Chinese Interface
• English Interface
• Japanese Interface

• Add emotion control features

• Framework: Next.js 14
• Language: TypeScript
• Styling: TailwindCSS
• UI Components: Radix UI
• State Management: Jotai
• Form Handling: React Hook Form
• HTTP Client: ky
• i18n: next-intl
• Theming: next-themes
• Code Standards: ESLint, Prettier
• Commit Standards: Husky, Commitlint

1. Clone the project

git clone https://github.com/302ai/302_ai_voice_arena
cd 302_ai_voice_arena

2. Install dependencies

pnpm install

3. Configure environment

cp .env.example .env.local

Modify the environment variables in .env.local as needed.

4. Start development server

pnpm dev

5. Build for production

pnpm build
pnpm start

302.AI is an enterprise-oriented AI application platform that offers pay-as-you-go services, ready-to-use solutions, and an open-source ecosystem.✨

1. 🧠 Comprehensive AI capabilities: Incorporates the latest in language, image, audio, and video models from leading AI brands.
2. 🚀 Advanced application development: We build genuine AI products, not just simple chatbots.
3. 💰 No monthly fees: All features are pay-per-use, fully accessible, ensuring low entry barriers with high potential.
4. 🛠 Powerful admin dashboard: Designed for teams and SMEs - managed by one, used by many.
5. 🔗 API access for all AI features: All tools are open-source and customizable (in progress).
6. 💡 Powerful development team: Launching 2-3 new applications weekly with daily product updates. Interested developers are welcome to contact us.