2025/08/22 by Tim Sainburg, Asaf Zorea · 2 voices
Computer Science · #Speech and Audio Processing #Blind Source Separation Techniques #Music and Audio Processing
paper · pdf · doi:10.1038/s41598-025-13108-x
Extracting signals from noisy backgrounds is a fundamental problem in signal processing across a variety of domains. In this paper, we introduce Noisereduce, an algorithm for minimizing noise across a variety of domains, including speech, bioacoustics, neurophysiology, and seismology. Noisereduce uses spectral gating to estimate a frequency-domain mask that effectively separates signals from noise. It is fast, lightweight, requires no training data, and handles both stationary and non-stationary noise, making it both a versatile tool and a convenient baseline for comparison with domain-specific applications. We provide a detailed overview of Noisereduce and evaluate its performance on a variety of time-domain signals.