LocalVQE: An Ultra-Compact Real-Time Noise Removal Model

·2026.05.04 21:54

Key point

LocalVQE, an ultra-compact audio model that removes real-time echo and noise using about 1 million parameters, has been released.

Details

LocalVQE is an ultra-compact audio model designed to remove echo and noise in real time. It is characterized by operating very lightly using only about 1 million (1M) parameters.

Key features are as follows:

Real-time processing: Minimizes latency to effectively suppress echo and noise in real-time audio streams.
Ultra-lightweight design: The very small number of parameters allows it to run smoothly even on low-spec devices or edge devices.

A live demo is currently available via Hugging Face Spaces.

edge-ai audio-ai noise-cancellation local-vqe tiny-model

Read originalShare on Threads

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.