If you've made digital humans before, you know voice synthesis is a key step. Plenty of TTS tools exist, but many are paid or web-only—and web tools raise privacy concerns.
Today I'm introducing an open-source AI voice tool from NetEase Youdao: EmotiVoice. It synthesizes emotional speech, and its Plus edition EmotiVoice-Plus supports multi-character scripted dialogue—greatly widening the use cases.
EmotiVoice is a fast text-to-speech engine with Chinese and English voices and 2000+ simulated voices. With commands, it outputs speech with emotion—happy, sad, excited, angry and more.
A friendly web UI plus batch-generation APIs make it even more efficient.

The Plus upgrade centers on multi-character scripted dialogue—deepening its use in content creation, games and multimedia teaching.
For fast deployment, we ship a one-click offline package—download, unzip, run locally, no complex setup.

Launching is simple—run the included 一键启动.bat. It opens the local web UI (usually http://localhost:8501/), where you run everything in the browser.

Single-line or multi-character script synthesis both work through an intuitive UI—pick a voice, enter text, set emotion, then synthesize and preview or download right on the page.
Project: https://gitee.com/mirrors/EmotiVoice