面向创作者和开发者的 Gradio 语音工作台，集成多引擎 TTS、零样本声音克隆、Whisper 转写、人声分离与多语言翻译，一站式完成配音和字幕制作。
开源 AI 视频本地化配音工具，把 YouTube/Bilibili 视频自动做语音识别、字幕翻译、声音克隆和混音渲染，一站式生成多语言配音成片。
用机器学习给卡拉OK打分的应用，能分离人声并自动评分，适合聚会娱乐场景。
一个基于 Gradio + React 的 WebUI，把 GPT-SoVITS、CosyVoice、XTTSv2、Bark、MusicGen、RVC 等二十多种语音合成、音频生成与音乐生成模型整合到同一界面，省去逐个部署的麻烦。
…o drums, bass, vocals, guitar, piano and other (HTDemucs, with optional BS-RoFormer for cleaner vocals). The new full-screen stem studio plays all stems in sync with per-
…bilities in the main package - README : Documents Demucs (audio source separation) and Flare (GGUF/WASM) engines with usage examples - Tests : 62 passing tests across 7 s
…动 - 修复转写结束后崩溃的 bug 111 - 修复手动添加多个字幕后表格不能关闭的 bug - Demucs 新增非人声音轨合并输出 110 - 新增人声、其他音轨二分输出 - 字幕显示及编辑相关功能更新 - 字幕编辑：添加批量增减时间戳功能 - 字幕表格显示： - 添加持续时间过短的时间戳背景色提示功能 - 说话人 在表格中单独显示
… models available via the Download Center. - Full Demucs v1, v2, v3, & v4 compatibility. - Additional models and application patches can be downloaded via the "Settings"
…DX-Net model (included in the new package) - Full Demucs v1 & v2 backward compatibility - Ability to download additional models and application patches straight from the
…manual upgrade. - The following has been added: - Demucs v3 has been fully implemented into the GUI. This includes 2 new UVR-trained Demucs v3 models. - Additional advanc