Stream Talker Server [v1.0.0]

Changes in this release

Added

• Initial release of Stream Talker Server
• FastAPI-based TTS server with Qwen3-TTS models (0.6B, 1.7B)
• Dynamic model loading with automatic unloading
• Voice cloning with persistent caching
• Batch synthesis with ZIP archive output
• Multi-language support (10 languages)
• RESTful API for model, voice, and synthesis management
• Docker deployment with GPU support (CUDA 13.0)
• Health checks and environment introspection
• Configurable generation limits to prevent hangs

---

Docker Image:

docker pull virtualzer0/stream-talker-server:v1.0.0
docker pull virtualzer0/stream-talker-server:latest

Full Changelog: https://github.com/VirtualZer0/StreamTalkerServer/blob/main/CHANGELOG.md