v6.2

DioxusLabs/dioxusv6.2Nov 6, 2025by snakers4

AI Summary

This release prioritizes stability and quality improvements for the Voice Activity Detection (VAD) model, specifically targeting edge cases and rare audio inputs to ensure higher robustness.

Key Highlights

Reworked VAD training paradigm for better robustness
Significant quality improvements on unusual and child voices
Enhanced stability on Out-of-Distribution (OOD) and rare data
Improved handling of muted voices and phone calls
Fixed type hint for min_silence_at_max_speech parameter

New Features

Improved handling of muted voices and muted speech
Enhanced stability for unusual, rare, and unique data
Better performance on lower quality phone calls
Support for cartoon voices and child voices

Full Release Notes

- VAD training paradigm reworked;
- Overall slight quality improvement (no metrics update);
- Higher stability on OOD / rare / strange / unique data;
- Significant quality improvements on various known edge cases:
  - Unusual voices
  - Child voices
  - Cartoon voices
  - Muted voices
  - Muted speech
  - Lower quality phone calls

## What's Changed

* Fix type hint for min_silence_at_max_speech (float -> int) by @Purfview in https://github.com/snakers4/silero-vad/pull/714
* Adamnsandle by @adamnsandle in https://github.com/snakers4/silero-vad/pull/717
* Adamnsandle by @adamnsandle in https://github.com/snakers4/silero-vad/pull/719

**Full Changelog**: https://github.com/snakers4/silero-vad/compare/v6.1...v6.2