Skip to content

v1.2

Latest

Choose a tag to compare

@prakharsr prakharsr released this 15 Nov 19:45

馃幆 Key Improvements

Advanced Audio Quality Assurance

  • Multi-stage error detection prevents audio artifacts and quality issues
  • Automatic retry logic with parameter adjustment for failed generations
  • Audio quality analysis detects silence, clipping, repetition, and stretching
  • Duration outlier detection identifies abnormally slow generations
  • Improved token repetition detection prevents infinite audio loops

Migration Notes

  • Update dependencies: uv pip install -r requirements.txt
  • Review new environment variables in .env.sample

This release significantly improves audio quality, reliability, and debugging capabilities while maintaining full API compatibility. The advanced error detection system catches issues before they affect users, and the comprehensive logging helps with troubleshooting and parameter tuning.

Full Changelog: v1.1...v1.2