[KongchangAI]
Model

CallCleaner

A lightweight bioacoustic denoising model that transfers capability via knowledge distillation from the general-purpose audio separation foundation model SAM-Audio, with approximately 19 million parameters, requiring no detector, and capable of running in real time on a CPU.

Timeline (last 90 days)

Nov 10

CallCleaner is a denoiser with only 19 million parameters, detector-free, and runs at approximately 90 times real time on a laptop CPU.

Unverified50%
Nov 10

The researchers used knowledge distillation to transfer the capabilities of SAM-Audio to a small student model, CallCleaner.

Unverified50%
Nov 10

CallCleaner leaves 11 dB less noise around calls than Biodenoising, the leading publicly available neural bioacoustic denoiser.

Unverified50%
Nov 10

Among fast, CPU-available systems, CallCleaner occupies an advantageous position on the Pareto frontier of noise removal versus call fidelity, and requires no detector.

Unverified50%
Nov 10

The approach used in this paper requires only lightweight temporal hints to guide the separation model to focus on time periods when calls occur, without requiring precise frame-by-frame masks, thus reducing dependence on detection accuracy and keeping the system detector-free and lightweight.

Unverified50%
Nov 10

CallCleaner is designed for single-vocalizer recordings and supports fast, private local denoising without specialized hardware.

Unverified50%
Nov 10

The researchers have open-sourced the code and weights of CallCleaner.

Unverified50%

All Facts (7)

Source Articles