Project

WhispSkrid is open source, licensed under the MIT License — see LICENSE in the repository.

Contributions are welcome: read the contribution guide and open an issue on the issue tracker before sending a large change, so the direction can be agreed first.

The project is currently at v1.0.1 — first stable release.

Roadmap

• The four supported languages (French, English, German, Spanish) are kept at parity, both in the interface and the documentation.
• The native packages (.deb, .rpm, .pkg.tar.zst) and the source install are maintained in parallel.
• The engine is faster-whisper (CTranslate2). A second backend (whisper.cpp) is considered for later, behind the same interface.

Credits

WhispSkrid is the result of a human-LLM collaboration. The roles are written as they were actually played; the full list is in CONTRIBUTORS.md.

• Ronan Davalan — architect and arbiter: vision, decisions, validation and testing.
• Claude Code (Anthropic) — systems engineer and lead developer: principal author of the code, the packaging, the tooling, the documentation and this website.
• Gemini (Google) — synthesizer and strategic advisor: architectural analysis and cross-validation of decisions.
• Muse Spark 1.3 — usage tester: command-line trials, .deb installation, reading of the documentation in four languages.
• DeepSeek V4 Flash — tester and reviewer: installation on a Raspberry Pi, review of documentation and website.
• Grok (xAI) — external validator: critical review and editorial audit of the PDF manuals.
• faster-whisper (SYSTRAN, CTranslate2) and the OpenAI Whisper models — the speech engine.