kokoroi-rs is an open-source Chinese text-to-speech engine rewritten in Rust, using ONNX Runtime for efficient inference. It supports over 50 multilingual voices, offers CLI and HTTP API interfaces, and achieves real-time synthesis rates. A key feature is zero-dependency static compilation, enabling easy deployment without runtime dependencies. This project demonstrates Rust's suitability for performance-critical audio applications and provides a practical alternative to cloud-based TTS services. For developers building voice interfaces or accessibility tools, kokoroi-rs offers a self-hosted, high-performance solution. Its open-source nature allows customization and integration into various workflows, from embedded systems to web services. The project's focus on Chinese TTS fills a gap in the open-source ecosystem, and its architecture could serve as a template for other languages.
An open-source Rust-based TTS engine for Chinese with 50+ voices, zero-dependency deployment, and real-time performance.