Speech enhancement based on spectral subtraction in wavelet domain

Hassani, Malihe; Mollaei, M.R. Karami

doi:10.1109/cspa.2011.5759904

Cited by 11 publications

(12 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Spectral subtraction method is one of the most popular methods for reducing the background noise [2] . In this method only an estimate of noise is needed for speech enhancement.…”

Section: (B) Spectral Subtraction Methodsmentioning

confidence: 99%

“…This estimate of noise spectrum is subtracted from the noisy speech spectrum to obtain cleaned speech at the output. Wavelet thresholding is another method of denoising the speech signal but it is not successfully implemented for background noise [2]reduction. In this paper a new hybrid method of speech enhancement is proposed in which first we take the wavelet transform of noisy speech signal that yields Approximation and detail coefficients at level 1.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Novel Hybrid System for Speech Enhancement In Wavelet Domain

Pandey¹,

Jaiswal²,

Agrawal³

2014

IOSRJECE

View full text Add to dashboard Cite

“…Spectral subtraction method is one of the most popular methods for reducing the background noise [2] . In this method only an estimate of noise is needed for speech enhancement.…”

Section: (B) Spectral Subtraction Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Novel Hybrid System for Speech Enhancement In Wavelet Domain

Pandey¹,

Jaiswal²,

Agrawal³

2014

IOSRJECE

View full text Add to dashboard Cite

“…The effect of far-end noise on speech signal can be tackled by using traditional noise suppression algorithms like minimum mean-square error (MMSE), short-time spectral amplitude (STSA) estimator [18], spectral subtraction methods [20], etc. The approaches proposed for far-end noise reduction techniques discussed in the literature [18][19][20] are not suitable in the present context as they focus on mitigating noise at the speaker end rather than at the receiver end.…”

Section: Introductionmentioning

confidence: 99%

“…The approaches proposed for far-end noise reduction techniques discussed in the literature [18][19][20] are not suitable in the present context as they focus on mitigating noise at the speaker end rather than at the receiver end. Near-end noise cannot be influenced because the listener is located in a noisy environment, and the noise reaches the ears with hardly any possibility to suppress [8].…”

Section: Introductionmentioning

confidence: 99%

Dominant Frequency Enhancement of Speech Signal to Improve Intelligibility and Quality

Premananda¹,

B.V.²

2015

IJIGSP

View full text Add to dashboard Cite

Abstract-In mobile devices, perceived speech signal deteriorates significantly in the presence of near-end noise as the signal arrives directly at the listener's ears in a noisy environment. There is an inherent need to increase the clarity and quality of the received speech signal in noisier environment. It is accomplished by incorporating speech enhancement algorithms at the receiver end. The objective is to improve the intelligibility and quality of the speech signal by dynamically enhancing the speech signal when the nearend noise dominates. This paper proposes a speech enhancement approaches by inculcating the threshold of hearing and auditory masking properties of the human ear. Incorporating the masking properties, the speech samples that are audible can be obtained. In low SNR environments, selective audible samples can be enhanced to improve the clarity of the signal rather than enhancing every loud sample. Intelligibility and quality of the enhanced speech signal are measured using Speech Intelligibility Index and Perceptual Evaluation of Speech Quality. Experimental results connote the intelligibility and quality improvement of the speech signal with the proposed method over the unprocessed far-end speech signal. This approach is efficient in overcoming the deterioration of speech signals in a noisy environment.

show abstract

“…The far-end noise scenario is tackled by traditional noise suppressing algorithms like minimum mean-square error (MMSE) short-time spectral amplitude (STSA) estimator [1], adaptive filtering algorithms [2,3], spectral subtraction methods [4,5], etc. For near-end noise, signal cannot be influenced because the listener is located in the noisy environment and the noise reaches the ears with hardly any possibility to intercept [9].…”

Section: Introductionmentioning

confidence: 99%

Near-End Perception Enhancement using Dominant Frequency Extraction

2014

IJAERD

View full text Add to dashboard Cite

Abstract-In general voice communication, speech is always affected by background noise which reduces perception quality of the speech signal. The incoming speech from far end speaker and near end noise reaches our ear at the same time, noise cannot be cancelled. Hence speech signal is enhanced with respect to variation in the near end noise. In this work, we propose two speech enhancing methods in order to combat near end noise. In first method, suitable gain is derived and smoothened for speech enhancement using weighted hearing curve and overall speech is enhanced to give better quality and perception. In second method we determine dominant frequencies components of speech signal and enhance only those frequencies instead enhancing the entire signal. The energy needed for enhancing is reduced in later approach when compared to former one. So a method of energy conservation is achieved. Quality o f enhanced speech signals is measured using perception evaluation of speech quality (PESQ) which is an ITU recommended standard for speech quality assessment.

show abstract

Speech enhancement based on spectral subtraction in wavelet domain

Cited by 11 publications

References 11 publications

A Novel Hybrid System for Speech Enhancement In Wavelet Domain

A Novel Hybrid System for Speech Enhancement In Wavelet Domain

Dominant Frequency Enhancement of Speech Signal to Improve Intelligibility and Quality

Near-End Perception Enhancement using Dominant Frequency Extraction

Contact Info

Product

Resources

About