Hey everyone, I built NOVA-VAD, a lightweight explainable Voice Activity Detector that outperforms every major open source VAD on real-world noisy audio. GitHub: https://github.com/monishmal3375/nova-vad

Short summary

A developer released NOVA-VAD, an open-source Voice Activity Detector designed to outperform existing solutions like Silero and WebRTC on real-world noisy audio. The tool is lightweight, requires no GPU, and emphasizes explainability. Available on GitHub for immediate integration.

• •Lightweight VAD that works without GPU
• •Outperforms Silero and WebRTC on noisy audio
• •Open-source on GitHub with emphasis on explainability

Generated with AI, which can make mistakes.