Noise estimation by minima controlled recursive averaging for robust speech enhancement

By Israel Cohen, Baruch Berdugo. IEEE Signal Processing Letters, 2002

604 citations recorded by the source corpus.

Subfields: Computational Mechanics, Signal Processing

View Paper: https://www.semanticscholar.org/paper/4a7eb0e7d2a01196eab0e4fb991be76dfa3e4c21

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

1. Ross Cutler
2. Chandan K. A. Reddy
3. P. Loizou
4. Harishchandra Dubey
5. Sebastian Braun
6. Kuldip K. Paliwal
7. Aaron Nicolson
8. Dong Yu

In Signal Processing

1. Dong Yu
2. Li Deng
3. Jinyu Li
4. Yifan Gong
5. Deliang Wang
6. Ross Cutler
7. Jasha Droppo
8. A. Acero

Citations in the corpus

• Robust Speech Recognition Using a Cepstral Minimum-Mean-Square-Error-Motivated Noise Suppressor (2008, methodology): “In our current implementation, the noise variance is estimated using a minimum-controlled recursive moving- average noise tracker similar to the one described in [1].”
• A noise-estimation algorithm for highly non-stationary environments (2006, methodology): “The local minimum in (Cohen, 2002) was found by tracking the minimum of noisy speech over a search window spanning L frames.”
• A Novel Framework and Training Algorithm for Variable-Parameter Hidden Markov Models (2009, methodology): “A minimum-controlled recursive moving-average noise tracker [4] was used in our system to track the noise power with the same procedure and parameters used in our MFCC-MMSE noise suppresser work reported in [29].”
• rVAD: An Unsupervised Segment-Based Robust Voice Activity Detection Method (2019, methodology): “In the classical speech enhancement framework [49], accurate noise estimation is important and the widely used methods include minimum statistics noise estimation (MSNE) [48], minimum mean-square error (MMSE) [50, 51], and minima controlled recursive averaging (MCRA) [52].”
• Speech enhancement with weighted denoising auto-encoder (2013, methodology): “Then, the noise power spectrum is estimated by Minima Controlled Recursive Averaging (MCRA) method [8].”
• A minimum-mean-square-error noise reduction algorithm on Mel-frequency cepstra for robust speech recognition (2008, methodology): “2 Estimation of ( ) and ( ) In our current implementation, the noise variance 2( ) is estimated using a minimum controlled recursive moving-average noise tracker similar to the one described in [1].”
• Weighted Speech Distortion Losses for Neural-Network-Based Real-Time Speech Enhancement (2020, background): “Although most classical SE methods have to accommodate their approaches [4, 5, 6, 7] for fulfilling causality, many DNN-based methods in the literature [3, 8, 9] do not enforce this constraint.”
• DeepMMSE: A Deep Learning Approach to MMSE-Based Noise Power Spectral Density Estimation (2020, methodology): “Another group of noise PSD tracking methods are time-recursive averaging algorithms, including the minima controlled recursive averaging (MCRA) method [15], the improved MCRA method (IMCRA) [16], and MCRA-2 [17].”