Repolode
Предложить
HAKORADev

HAKORADev/VODER

Local, offline Python toolkit that combines speech-to-text, text-to-speech, voice cloning and conversion, music generation, and audio cleanup in one CLI and GUI.

Open source альтернатива:ElevenLabsDescript

Открыть на GitHub
Звёзды
523
Форки
92
Открытые issues
1
Последний коммит
18 дней назад

Что это?

VODER is a voice processing engine that bundles several audio workflows under one interface: speech-to-text, text-to-speech with voice design and cloning, voice conversion, music generation and stem separation, sound enhancement, sound effect generation, vocal separation, and speaker diarization. It runs on the user's own machine without a subscription and can operate on CPU or GPU.

Beyond single modes, it supports multi-speaker dialogue scripts with per-line timing and sound effects, language dubbing that keeps the original speaker's voice, and video input and output for voice conversion. Users can save voice clones for reuse, run utility tasks such as URL download and audio editing, and chain multiple tasks into named pipelines.

The project also includes an optional add-on called Project Eva for image, video, 3D scene generation and a local chat model. Setup requires FFmpeg, and some modes need a Hugging Face token and substantial VRAM or RAM. The author has stated the project is in maintenance mode with no further major feature development.

Ключевые возможности

  • 01Speech-to-text transcription with translation and speaker diarization
  • 02Text-to-speech with voice design from text descriptions and voice cloning from reference clips
  • 03Voice conversion for audio and video, including music at 44.1kHz
  • 04Music generation from lyrics and style prompts, with separate instrument stems
  • 05Vocal separation, speaker separation, and sound enhancement with 48kHz upscaling
  • 06Multi-speaker dialogue scripts with sound effects and automatic background music

Сценарии использования

  • Dubbing a video into another language while keeping the original speaker's voice
  • Separating individual speakers from a meeting or interview recording
  • Cloning a narrator's voice once and reusing it across many TTS projects
  • Isolating vocals or instrumentals from songs for remixing

Темы

#voice-cloning#text-to-speech#speech-to-text#voice-conversion#music-generation#audio-processing#speaker-diarization#offline-ai#tts#voder

Частые вопросы

Что такое VODER?
Local, offline Python toolkit that combines speech-to-text, text-to-speech, voice cloning and conversion, music generation, and audio cleanup in one CLI and GUI. VODER is a voice processing engine that bundles several audio workflows under one interface: speech-to-text, text-to-speech with voice design and cloning, voice conversion, music generation and stem separation, sound enhancement, sound effect generation, vocal separation, and speaker diarization. It runs on the user's own machine without a subscription and can operate on CPU or GPU.
VODER можно использовать бесплатно? Какая у него лицензия?
VODER распространяется с открытым исходным кодом под лицензией AGPL-3.0. Перед использованием в коммерческих продуктах проверьте условия лицензии.
VODER всё ещё поддерживается?
Активно поддерживается. Последний коммит в VODER — 1 окт. 2026 г..
Для чего можно использовать VODER?
Dubbing a video into another language while keeping the original speaker's voice Separating individual speakers from a meeting or interview recording Cloning a narrator's voice once and reusing it across many TTS projects Isolating vocals or instrumentals from songs for remixing
Где скачать VODER?
Исходный код и инструкции по установке — на GitHub: https://github.com/HAKORADev/VODER

Описание создано ИИ на основе README проекта. Статистика GitHub обновляется ежедневно.

Похожие проекты