DubbiOvi 2.0 is a free, local-first platform for audiovisual translation, dubbing, speech synthesis and timeline synchronization, designed for educators, researchers and translation professionals.
Powerful tools that simplify translation, speech synthesis, and dubbing projects.
Transcription, voice synthesis, and timeline projects run directly on your workstation. Local-first storage guarantees local data control, offline capability, and native .dubbiovi project files.
Run speech recognition locally with downloadable Whisper runtimes and models for private offline execution, or connect optional cloud transcription services for additional flexibility.
Generate speech with local Piper neural voices for offline synthesis, or use optional cloud providers like Google Gemini and OpenAI-compatible endpoints. Create, regenerate, and audition audio takes with ease.
Audio waveform timeline, dedicated Playback Mixer, and interactive Dub Preview. Synchronize speech segments, adjust volume levels, and fit voice timing with visual feedback.
Render synchronized dubs directly to final video files, export individual speech takes, mixed audio, and structured bilingual script tables (Word, Excel, CSV, JSON).
Free and open under EUPL-1.2. Designed for macOS and Windows with support for offline local runtimes as well as optional cloud AI providers.
A streamlined, 5-step process tailored for professional audiovisual translators. Click the steps to view details.
Real preview of the DubbiOvi 2.0 translation and dubbing studio workspace, voice synthesis configuration, and media exporting tools.
Central dubbing workspace featuring video playback, waveform timeline visualization, take management, and Playback Mixer.
Configure local Whisper models, Piper TTS voice models, and optional Gemini or OpenAI-compatible AI provider keys.
Export finalized dubs to complete video containers, mixed audio, project files, and structured script formats.
DubbiOvi 2.0 was created and developed by Alfonso C. Rodríguez Fernández-Peña.
The software serves as a modern platform for teaching audiovisual translation techniques, facilitating speech-processing research datasets, and transferring AI dubbing technologies directly into educational and professional workflows. By enabling local control over proprietary keys, local models, and offline execution, the project aims to foster open translation studies globally.
Download the platform and start building local-first translation workflows.
Download / Register