v6.2

voideditor/voidv6.2Nov 6, 2025by snakers4

AI Summary

This release focuses on reworking the VAD training paradigm to improve overall stability and quality, specifically targeting difficult audio scenarios like muted voices and phone calls.

Key Highlights

VAD training paradigm reworked
Higher stability on out-of-distribution and rare data
Quality improvements on unusual, muted, child, and cartoon voices
Type hint fix for min_silence_at_max_speech parameter

Breaking Changes

Type hint for min_silence_at_max_speech changed from float to int

New Features

Better performance on lower quality phone calls
Enhanced stability on strange and unique data
Improved handling of muted speech and 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