v0.8.9 v0.8.10
low risk
ysharma3501/NovaSRSep 26, 2026 → Sep 27, 20260 days between releases
What's New
Adds CPU core splitting and improves streaming startup with logging optimizations
New Features
- New opt-in `--cpu-core-split` flag for Windows physical core optimization
- New `--demo-voice` cloning example for CPU and GPU servers
- Server logging runs outside the request event loop
- Voice Studio defaults to 0 ms additional browser buffering
Improvements
- CPU synthesis continues through initial quiet audio without retries
- CPU native version bumped to 0.4.2
- Voice Studio prepares and reuses audio context before requests
Bug Fixes
- Removes CPU quiet-onset retries to prevent synthesis interruptions
Upgrade Notes
CPU native version updated from 0.4.1 to 0.4.2. Windows users can enable `--cpu-core-split` to utilize physical cores for better performance. Voice Studio buffering defaults changed to 0 ms; adjust playback options if needed.
The Two Releases
v0.8.9
RealtimeTTS 0.8.9
Sep 26, 2026· 25 days ago
Makes Qwen CPU and GPU servers easier to install and use, with corrected streaming playback in Voice Studio.
Full release notesv0.8.10
RealtimeTTS 0.8.10
Sep 27, 2026· 25 days ago
Improves Qwen streaming startup and removes CPU quiet-onset retries. Enhances Voice Studio defaults and server logging performance.
Full release notes