Voice Activity Detection with Generalized Gamma Distribution

Almpanidis, George; Kotropoulos, Constantine

doi:10.1109/icme.2006.262692

Cited by 3 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Speech presence probability (SPP) values have been widely used to derive minimum mean-square (MMSE) soft-decision spectral speech enhancement algorithms ( [1]- [3]), which can be utilized to improve perceptual quality or noise robust automatic speech recognition (ASR) rates. Additionally, statistical model-based voice activity detection (VAD) algorithms typically utilize a theory related to SPPs when determining active speech regions ( [6], [17], [18]). …”

Section: Introductionmentioning

confidence: 99%

Improved Speech Presence Probabilities Using HMM-Based Inference, With Applications to Speech Enhancement and ASR

Borgström

Alwan

2010

IEEE J. Sel. Top. Signal Process.

View full text Add to dashboard Cite

Abstract-This paper presents a technique for determining improved speech presence probabilities (SPPs), by exploiting the temporal correlation present in spectral speech data. Based on a set of traditional SPPs, we estimate the underlying speech presence probability via statistical inference. Traditional SPPs are assumed to be observations of channel-specific two-state Markov models. Corresponding steady-state and transitional statistics are set to capture the well-known temporal correlation of spectral speech data, and observation statistics are modeled based on the effect of additive acoustic noise on resulting SPPs. Once underlying models have been parameterized, improved speech presence probabilities can be estimated via traditional inference techniques, such as the forward or forward-backward algorithms. The 2-state configuration of underlying signal models enables low complexity HMM-based processing, only slightly increasing complexity relative to standard SPPs, and thereby making the proposed framework attractive for resource-constrained scenarios. Proposed SPP masks are shown to provide a significant increase in accuracy relative to the state-of-the-art method of [12], in terms of the mean pointwise Kullback-Leibler (KL) distance. When applied to soft-decision speech enhancement, proposed SPPs show improved results in terms of segmental SNRs. Closer analysis reveals significantly decreased noise leakage, whereas speech distortion is increased. When applied to automatic speech recognition (ASR), the use of soft-decision enhancement with proposed SPPs provides increased recognition performance, relative to [12].

show abstract

Section: Introductionmentioning

confidence: 99%

Improved Speech Presence Probabilities Using HMM-Based Inference, With Applications to Speech Enhancement and ASR

Borgström

Alwan

2010

IEEE J. Sel. Top. Signal Process.

View full text Add to dashboard Cite

show abstract

Convex Combination of Multiple Statistical Models With Application to VAD

Petsatodis

Boukis

Talantzis

et al. 2011

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Dimensionality Reduction in Feature Vector using Principle Component Analysis (PCA) for Effective Speaker Recognition

BabuKorada¹,

Anitha.²,

S³

2013

IJAIS

View full text Add to dashboard Cite

This paper describes analysis of a speaker recognition model based on Generalized Gamma Distribution (GGD) using PCA. The proposed work mainly concentrates on the feature vectors that are generated from the speech signals contain high dimension data, but to model a speech and recognize a speaker finite speech samples which plays significant role in speech analysis are sufficient, hence it necessary to reduce dimension of the data. The PCA is considered for this purpose, it converts high dimension speech signal in to a low dimension speech signal by transforming the un-correlated components of the speech signal. PCA not only reduces the correlation among feature vectors but also the speech signal. The feature vectors are modeled by extracting MFCC followed by PCA for dimensionality reduction.

show abstract

Voice Activity Detection with Generalized Gamma Distribution

Cited by 3 publications

References 14 publications

Improved Speech Presence Probabilities Using HMM-Based Inference, With Applications to Speech Enhancement and ASR

Improved Speech Presence Probabilities Using HMM-Based Inference, With Applications to Speech Enhancement and ASR

Convex Combination of Multiple Statistical Models With Application to VAD

Dimensionality Reduction in Feature Vector using Principle Component Analysis (PCA) for Effective Speaker Recognition

Contact Info

Product

Resources

About