无需训练数据的通用时序信号降噪算法，适用于语音、生物声学等多领域。
Noisereduce: Domain General Noise Reduction for Time Series Signals
- 基于频域谱门控估计噪声掩码，实现信号与噪声分离。
- 处理稳态与非稳态噪声，速度快且资源消耗低。
- 适合需要快速降噪基准或跨领域应用的研究者使用。
从噪声背景中提取信号是多个领域信号处理中的基础问题。本文提出Noisereduce，一种适用于语音、生物声学、神经生理学和地震学等多种领域的通用降噪算法。该方法利用频域谱门控技术，估计频率域掩码以有效分离信号与噪声。Noisereduce具有计算速度快、模型轻量、无需训练数据、可处理稳态与非稳态噪声的特点，既是一个多功能工具，也可作为领域特定方法的便捷对比基线。我们对Noisereduce在多种时序信号上的性能进行了详细评估。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。