VOICEVOX is a text-to-speech editor that provides free, medium-quality voice synthesis with a Japanese-language focus.
The tool addresses the need for accessible speech synthesis by pairing a TypeScript and Vue-based editor frontend with a separate engine and core component. Users compose text and select from available voices, with the editor handling the interface while delegating synthesis work to the VOICEVOX ENGINE backend. The architecture separates concerns across three main components: the editor, the engine, and the core synthesis library.
The editor targets users seeking free text-to-speech synthesis without licensing restrictions. It suits projects requiring Japanese voice output or developers building applications that integrate speech synthesis. The tool is designed for both standalone use and as a component within larger systems, with the engine available as a separate service that the editor communicates with. Adoption requires running the VOICEVOX ENGINE locally or pointing the editor to an engine instance; Windows users typically reference the bundled engine at a standard installation path, while macOS users specify the engine location within the application bundle.
The project actively welcomes external contributions beyond code, explicitly encouraging documentation writing, test generation, and design discussion. Development coordination uses issue tracking to prevent duplicate work, with contributors expected to signal their intent before starting. The codebase is acknowledged as complex due to its use of Electron, TypeScript, Vue, and Vuex, and the maintainers provide architectural documentation and code walkthroughs to help new contributors navigate the structure. An unofficial Discord server serves as a space for development discussion and community interaction.
