Public
Star 历史趋势
数据来源: GitHub API · 生成自 Stargazers.cn
README.md
English | 简体中文
AudioScript
Listen to your system audio and turn foreign podcasts, videos, and meetings into live bilingual subtitles: source text plus translation. Recognition and translation run entirely on your machine, with 29+ languages supported. On macOS 26+, system translation is also available.
https://github.com/user-attachments/assets/b1f88a82-b3fe-4721-8736-af6a9e4d828e
Features
- Capture: system audio / microphone / both mixed.
- Background: tray-resident, closing the window does not quit, single instance, launch at login.
- Recognition: bundled Zipformer streaming bilingual recognition; optional X-ASR, Nemotron 3.5, and Qwen3-ASR.
- Translation: bundled Opus-MT en ↔ zh; optional LMT-60 and NLLB-200.
- Overlay: bilingual source and translation; adjustable font size, opacity, line count, and a colorful flowing border.
- Internationalization: zh-CN / zh-TW / en / ja / ko / fr / de / es / pt-BR / ru / it / id / ar.
Privacy
- Speech recognition and translation run entirely on-device and work offline.
- Anonymous usage analytics via Aptabase and crash reports via Sentry; disable both under General → Privacy.
Build
pnpm install # deps + ASR model and Silero VAD (≈ 59 MB)
pnpm fetch:mt # fetch models
pnpm tauri dev # dev mode
pnpm tauri build # packageContact
- X: @niyongsheng
- E-mail: niyongsheng@Outlook.com
License
-
sherpa-onnx Apache-2.0, X-ASR Apache-2.0, Nemotron 3.5 ASR OpenMDW-1.1 (optional), Qwen3-ASR Apache-2.0, Silero VAD MIT, CTranslate2 MIT, opus-mt, LMT-60 Apache-2.0 (optional), NLLB-200 CC-BY-NC-4.0 (optional), translation-rs MIT OR Apache-2.0, flexaudio MIT, tauri-nspanel MIT OR Apache-2.0