Legal
Licensing
Draft — plain-language summary, not legal advice. Final wording is owner-reviewed.
The application code
Castwright is published under the Functional Source License 1.1 with an Apache-2.0 future grant (FSL-1.1-Apache-2.0).
In plain language: the source code goes public with the beta — you will be able to read it, run it, and fork it for personal use. What is restricted for two years after each release: redistribution for competing commercial purposes. Two years after any release, that release converts to Apache-2.0 and the restriction lifts permanently.
Source-available is not the same as OSI open source, and we say so plainly rather than obscuring it. The FSL is purpose-built for exactly this situation: protecting a small team's ability to sustain the project while keeping the code public and the community able to audit, learn from, and build on it. If that tradeoff isn't for you, that's a fair position — we'd rather be honest about it than pretend otherwise.
The full licence text is in the LICENSE file in the GitHub repository (github.com/dudarenok-maker/Castwright, public with the beta).
Brand assets
The Castwright name, the Castwave mark (the waveform-and-book logo), and associated visual identity are all rights reserved. They are not covered by the FSL. See /legal/brand for fair-use guidance and what's permitted.
Engine licences
Castwright bundles or downloads several open-source speech engines. Their licences apply to their respective models and code:
- Kokoro — Apache-2.0. Weights ship with the installer.
- Qwen TTS — Apache-2.0 (verify per release; licence may change with model versions).
- Coqui XTTS v2 — Coqui Public Model Licence (CPML), non-commercial use only. Castwright downloads XTTS weights on demand at your request; you supply them and are responsible for complying with the CPML.
Castwright stands on open-source speech engines — credited by name:
- Kokoro — fast, high-quality English voices (Apache-2.0)
- Coqui XTTS — voice cloning from a short sample (CPML; download-on-demand, user-supplied)
- Qwen — voice design: build a character's voice from a description (Apache-2.0; verify per release)