Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog ⧉, and this project adheres to Semantic Versioning ⧉.

0.2.0 - Unreleased

Security

• GHSA-w476-p2h3-79g9 ⧉ determined to not be a vulnerability in aquarion-libtts itself. Only the development environment is impacted, and even there, the risk is low. This will be fixed in the near future.

Added

• Support for Python 3.13.

Fixed

• Nothing yet.

Changed

• Replaced Sphinx with Mkdocs for documentation generation.
• BREAKING CHANGE: Renamed KokoroDeviceNames to KokoroDeviceTypes.
• BREAKING CHANGE: TTSAudioSpec.format attribute changed to mime_type.
• BREAKING CHANGE: KokoroBackend now audio output now conforms to the audio/L16 MIME type, which means it’s output is now in big endian byte order instead of the previous little endian.

Deprecated

• Nothing yet.

Removed

• Nothing yet.

---

0.1.1 - 2025-10-23

Fixed

• Fixed en-core-web-sm dependency failure when trying to pip install aquarion-libtts.
• Fixed TTSPluginRegistry.load_plugins not working without the kokoro extra.

---

0.1.0 - 2025-10-17

Security

• CVE-2025-8869 ⧉ determined to not be a risk because the minimum version Python version of aquarion-libtts is 3.12.

Added

• Core project structure and artifacts.
• Dev environment & tools setup.
• Initial user stories and acceptance criteria.
• A plugin system for TTS backends.
• Internationalization and localization support.
• Support for English and French languages.
• Kokoro TTS as the first text-to-speech backend that is supported.
• A settings specification and description system for dynamic settings UI generation.
• Full documentation.

Fixed

• Nothing, first release.

Changed

• Nothing, first release.

Deprecated

• Nothing, first release.

Removed

• Nothing, first release.