Noise estimation by minima controlled recursive averaging for robust speech enhancement

Cohen, Israel; Berdugo, Baruch

doi:10.1109/97.988717

Cited by 488 publications

(288 citation statements)

References 6 publications

Supporting

Mentioning

286

Contrasting

Unclassified

Order By: Relevance

“…In addition, waveform plots and spectrogram observation were also conducted for performance evaluation. The original MCRA noise estimator [3], the forward-backward MCRA (FB_MCRA) noise estimator [8] were conducted for performance comparison. A three-step-decision gain factor [16] was employed to perform speech denoising for various noise estimators.…”

Section: Resultsmentioning

confidence: 99%

“…In (7), the threshold of speech-presence δ γ is a constant in the MCRA algorithm [3]. If the value of δ γ is too high, a greater quantity of weak speech spectra, such as weak vowels and consonants, would be classified as noise.…”

Section: Speech Presence Probability Adapted By Harmonic Propertiesmentioning

confidence: 99%

“…The minimum P min (η, Ω) is employed to determine the value of the speech indicator given in (7) and (8) for the MCRA method [3].…”

Section: Of 22mentioning

confidence: 99%

“…Many studies have been conducted to estimate noise [1][2][3][4][5][6][7][8][9][10][11]. Kianyfar and Abutalebi [1] proposed a noise estimator, which employed speech presence probability to update noise variance.…”

Section: Introductionmentioning

confidence: 99%

“…They are presented as [3] H 0 (η, where S(η, Ω) and D(η, Ω) represent the spectrum of clean speech and additive noise, respectively.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Estimation of Noise Magnitude for Speech Denoising Using Minima-Controlled-Recursive-Averaging Algorithm Adapted by Harmonic Properties

Lei

Shen

et al. 2016

Applied Sciences

View full text Add to dashboard Cite

Abstract:The accuracy of noise estimation is important for the performance of a speech denoising system. Most noise estimators suffer from either overestimation or underestimation on the noise level. An overestimate on noise magnitude will cause serious speech distortion for speech denoising. Conversely, a great quantity of residual noise will occur when the noise magnitude is underestimated. Accurately estimating noise magnitude is important for speech denoising. This study proposes employing variable segment length for noise tracking and variable thresholds for the determination of speech presence probability, resulting in the performance improvement for a minima-controlled-recursive-averaging (MCRA) algorithm in noise estimation. Initially, the fundamental frequency was estimated to determine whether a frame is a vowel. In the case of a vowel frame, the increment of segment lengths and the decrement of threshold for speech presence were performed which resulted in underestimating the level of noise magnitude. Accordingly, the speech distortion is reduced in denoised speech. On the contrary, the segment length decreases rapidly in noise-dominant regions. This enables the noise estimate to update quickly and the noise variation to track well, yielding interference noise being removed effectively through the process of speech denoising. Experimental results show that the proposed approach has been effective in improving the performance of the MCRA algorithm by preserving the weak vowels and consonants. The denoising performance is therefore improved.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Speech Presence Probability Adapted By Harmonic Propertiesmentioning

confidence: 99%

“…The minimum P min (η, Ω) is employed to determine the value of the speech indicator given in (7) and (8) for the MCRA method [3].…”

Section: Of 22mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…They are presented as [3] H 0 (η, where S(η, Ω) and D(η, Ω) represent the spectrum of clean speech and additive noise, respectively.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Estimation of Noise Magnitude for Speech Denoising Using Minima-Controlled-Recursive-Averaging Algorithm Adapted by Harmonic Properties

Lei

Shen

et al. 2016

Applied Sciences

View full text Add to dashboard Cite

show abstract

Feature Compensation

Droppo

2012

Techniques for Noise Robustness in Automatic Speech Recognition

View full text Add to dashboard Cite

ASR Models from Conventional Statistical Models to Transformers and Transfer Learning

Sherly,

Pillai,

Manohar

2024

Automatic Speech Recognition and Translation for Low Resource Languages

View full text Add to dashboard Cite

Noise estimation by minima controlled recursive averaging for robust speech enhancement

Cited by 488 publications

References 6 publications

Estimation of Noise Magnitude for Speech Denoising Using Minima-Controlled-Recursive-Averaging Algorithm Adapted by Harmonic Properties

Estimation of Noise Magnitude for Speech Denoising Using Minima-Controlled-Recursive-Averaging Algorithm Adapted by Harmonic Properties

Feature Compensation

ASR Models from Conventional Statistical Models to Transformers and Transfer Learning

Contact Info

Product

Resources

About