Speech Absence Research Articles

Voice Activity Detection Using Global Speech Absence Probability Based on Teager Energy for Speech Enhancement

Voice Activity Detection Using Global Speech Absence Probability Based on Teager Energy for Speech Enhancement

On using acoustic environment classification for statistical model-based speech enhancement

In this paper, we present a statistical model-based speech enhancement technique using acoustic environment classification supported by a Gaussian mixture model (GMM). In the data training stage, the principal parameters of the statistical model-based speech enhancement algorithm such as the weighting parameter in the decision-directed (DD) method, the long-term smoothing parameter of the noise estimation, and the control parameter of the minimum gain value are uniquely set as optimal operating points according to the given noise information to ensure the best performance for each noise. These optimal operating points, which are specific to the different background noises, are estimated based on the composite measures, which are the objective quality measures representing the highest correlation with the actual speech quality processed by noise suppression algorithms. In the on-line environment-aware speech enhancement step, the noise classification is performed on a frame-by-frame basis using the maximum likelihood (ML)-based Gaussian mixture model (GMM). The speech absence probability (SAP) is used to detect the speech absence periods and to update the likelihood of the GMM. According to the classified noise information for each frame, we assign the optimal values to the aforementioned three parameters for speech enhancement. We evaluated the performances of the proposed methods using objective speech quality measures and subjective listening tests under various noise environments. Our experimental results showed that the proposed method yields better performances than does a conventional algorithm with fixed parameters.

환경잡음분류 기반의 향상된 음성부재확률 추정

본 논문에서는 음성향상을 위하여 환경잡음분류를 적용한 향상된 음성부재확률 추정방법을 제안한다. 기존의 음성부재확률 추정방법에서는 마이크로폰 입력신호와 추정된 잡음신호 기반의 a posteriori SNR값에 문턱값을 적용하여 음성부재확률을 구하는데 필요한 음성부재의 a priori 확률을 도출하였다. 본 논문에서 제안된 알고리즘은 보다 효과적인 음성부재확률 추정을 위하여 고정된 문턱값과 스무딩 (smoothing)파라미터를 사용하는 기존의 방법과는 달리 잡음분류 알고리즘인 가우시안 혼합 모델 (Gaussian mixture model)을 사용하여 잡음마다 최적화된 파라미터를 적용한다. 제안된 음성 향상 기법은 ITU-T P.862 PESQ (perceptual evaluation of speech quality)와 composite measure를 이용하여 다양한 환경에서 평가하였으며, 제안된 알고리즘이 기존의 음성부재확률 추정방법보다 향상된 결과를 보였다. In this paper, we propose a improved speech absence probability estimation algorithm by applying environmental noise classification for speech enhancement. The previous speech absence probability required to seek a priori probability of speech absence was derived by applying microphone input signal and the noise signal based on the estimated value of a posteriori SNR threshold. In this paper, the proposed algorithm estimates the speech absence probability using noise classification algorithm which is based on Gaussian mixture model in order to apply the optimal parameter each noise types, unlike the conventional fixed threshold and smoothing parameter. Performance of the proposed enhancement algorithm is evaluated by ITU-T P.862 PESQ (perceptual evaluation of speech quality) and composite measure under various noise environments. It is verified that the proposed algorithm yields better results compared to the conventional speech absence probability estimation algorithm.

잡음환경에서 Teager 에너지의 전력 스펙트럼 편차에 기반한 음성 검출 알고리즘

본 논문에서는 잡음환경에서 효과적인 음성을 검출하기 위한 새로운 음성검출 (VAD, voice activity detection) 알고리즘을 제안한다. 제안된 방법은 개선된 음성/비음성 검출을 위해 기존의 파워 스펙트럼 편차를 적용하는 대신 Teager 에너지 기반의 파워 스펙트럼 편차 (power spectral deviation)를 이용한다. 또한 향상된 VAD 성능을 위하여 각각의 주파수 밴드에 대한 음성부재확률 (speech absence probability)을 제안된 파워 스펙트럼 편차를 도출하는데 스무딩 (smoothing) 파라미터로 적용한다. 제안된 알고리즘은 기존의 방법과 객관적인 실험을 통해 비교 평가한 결과 다양한 배경잡음 환경에서 향상된 성능을 보였다. In this paper, we propose a novel voice activity detection (VAD) algorithm to effectively distinguish speech from nonspeech in various noisy environments. The presented VAD utilizes the power spectral deviation (PSD) based on Teager energy (TE) instead of the conventional PSD scheme to improve the performance of decision for speech segments. In addition, the speech absence probability (SAP) is derived in each frequency subband to modify the PSD for further VAD. Performances of the proposed VAD algorithm are evaluated by objective test under various environments and better results compared with the conventional methods are obtained.

Speech dereverberation in noisy environments using an adaptive minimum mean square error estimator

The authors present here a novel method for reducing the late reverberation of speech signals in noisy environments. In this method, the amplitude of clean signal is obtained by an adaptive estimator that minimises the mean square error (MSE) under signal presence uncertainty. The spectral gain function, that is an adaptive variable-order minimum MSE estimator, is obtained as a weighted geometric mean of hypothetical gains associated with speech presence and absence. The order of estimator is estimated for each time frame and each frequency component individually. The authors propose the adaptation of order of estimator according to the probability of speech presence, which makes the estimation more accurate. The evaluations confirm superiority of the proposed method in dereverberation of speech signals in noisy environments.

Adaptive microphone array processing for high-performance speech recognition in car environment

In this paper, a novel speech enhancement algorithm is proposed. The algorithm controls the amount of noise reduction according to whether the target speech is absence or presence in noisy environments. Based on the estimated speech absence probability (SAP), the amount of noise reduction is adaptively controlled. To calculate the SAP, normalized cross correlation of linear predictive residual signals instead of that of original input signals is utilized. It is especially robust and effective in reverberant and realistic environments. Experimental results show that the proposed algorithm improves speech recognition rates compared with conventional beamforming algorithms in car environments.

A Data-Driven Approach to A Priori SNR Estimation

The a priori signal-to-noise ratio (SNR) plays an important role in many speech enhancement algorithms. In this paper, we present a data-driven approach to a priori SNR estimation. It may be used with a wide range of speech enhancement techniques, such as, e.g., the minimum mean square error (MMSE) (log) spectral amplitude estimator, the super Gaussian joint maximum a posteriori (JMAP) estimator, or the Wiener filter. The proposed SNR estimator employs two trained artificial neural networks, one for speech presence, one for speech absence. The classical decision-directed a priori SNR estimator by Ephraim and Malah is broken down into its two additive components, which now represent the two input signals to the neural networks. Both output nodes are combined to represent the new a priori SNR estimate. As an alternative to the neural networks, also simple lookup tables are investigated. Employment of these data-driven nonlinear a priori SNR estimators reduces speech distortion, particularly in speech onset, while retaining a high level of noise attenuation in speech absence.

A single channel speech enhancement technique exploiting human auditory masking properties

Abstract. To enhance extreme corrupted speech signals, an Improved Psychoacoustically Motivated Spectral Weighting Rule (IPMSWR) is proposed, that controls the predefined residual noise level by a time-frequency dependent parameter. Unlike conventional Psychoacoustically Motivated Spectral Weighting Rules (PMSWR), the level of the residual noise is here varied throughout the enhanced speech based on the discrimination between the regions with speech presence and speech absence by means of segmental SNR within critical bands. Controlling in such a way the level of the residual noise in the noise only region avoids the unpleasant residual noise perceived at very low SNRs. To derive the gain coefficients, the computation of the masking curve and the estimation of the corrupting noise power are required. Since the clean speech is generally not available for a single channel speech enhancement technique, the rough clean speech components needed to compute the masking curve are here obtained using advanced spectral subtraction techniques. To estimate the corrupting noise, a new technique is employed, that relies on the noise power estimation using rapid adaptation and recursive smoothing principles. The performances of the proposed approach are objectively and subjectively compared to the conventional approaches to highlight the aforementioned improvement.

Minima-controlled speech presence uncertainty tracking method for speech enhancement

In speech enhancement, soft decision, in which the speech absence probability (SAP) is introduced to modify the spectral gain or update the noise power, is known to be efficient. In many previous works, a fixed a priori probability of speech absence ( q) is assumed in estimating the SAP, which is not realistic since speech is quasi-stationary and may not be present in each frequency bin. To address this problem, Malah et al. devised a novel method to obtain distinct values of q for each frequency bin in many frames by comparing the a posteriori SNR to a threshold value [9]. In this regard, a novel algorithm is achieved by taking an advantage of a minima-controlled recursive averaging (MCRA) technique that allows for the robust tracking of speech absence in time. This leads to the improved tracking performance of speech absence in speech enhancement and better results in the objective and subjective evaluation tests.

Probabilistic Adaptation Mode Control Algorithm for GSC-Based Noise Reduction

An efficient noise reduction algorithm is proposed to improve speech recognition performance for human machine interfaces. In the algorithm, a probabilistic adaptation mode controller (AMC) is designed and adopted to the generalized sidelobe canceller (GSC). To detect target speech intervals, the proposed AMC calculates the inter-channel correlation and estimates speech absence probability (SAP). Based on the SAP, the adaptation mode of the adaptive filter in the GSC is decided. Experimental results show the proposed algorithm significantly improves speech recognition performances and signal-to-noise ratios in real noisy environments.

Improved Global Soft Decision Incorporating Second-Order Conditional MAP in Speech Enhancement

In this paper, we propose a novel method based on the second-order conditional maximum a posteriori (CMAP) to improve the performance of the global soft decision in speech enhancement. The conventional global soft decision scheme is found through investigation to have a disadvantage in that the global speech absence probability (GSAP) in that scheme is adjusted by a fixed parameter, which could be a restrictive assumption in the consecutive occurrences of speech frames. To address this problem, we devise a method to incorporate the second-order CMAP in determining the GSAP, which is clearly different from the previous approach in that not only current observation but also the speech activity decisions of the previous two frames are exploited. Performances of the proposed method are evaluated by a number of tests in various environments and show better results than previous work.

Optimal Gain Filter Design for Perceptual Acoustic Echo Suppressor

This Letter proposes an optimal gain filter for the perceptual acoustic echo suppressor. We designed an optimally-modified log-spectral amplitude estimation algorithm for the gain filter in order to achieve robust suppression of echo and noise. A new parameter including information about interferences (echo and noise) of single-talk duration is statistically analyzed, and then the speech absence probability and the a posteriori SNR are judiciously estimated to determine the optimal solution. The experiments show that the proposed gain filter attains a significantly improved reduction of echo and noise with less speech distortion.

Global Soft Decision Employing Support Vector Machine For Speech Enhancement

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> In this letter, we propose a novel speech enhancement technique based on global soft decision incorporating a support vector machine (SVM). Global soft decision in the proposed approach is performed employing the probabilistic outputs of the SVM rather than the conventional Bayes' rule. Actually, global speech absence probability (GSAP) is determined by the sigmoid function based on key parameters estimated by the model-trust minimization algorithm of the SVM output. Improved results are obtained in terms of speech quality measures for various types of noise and at different signal-to-noise ratio (SNR) levels when the proposed SVM is adopted in the global soft decision for speech enhancement. </para>

Speech Reinforcement Based on Soft Decision under Far-End Noise Environments

In this letter, we propose a speech reinforcement technique based on soft decision under both the far-end and near-end noise environments. We amplify the estimated clean speech signal at the far-end based on the estimated ambient noise spectrum at the near-end, as opposed to reinforcing the noisy far-end speech signal, so that it can be heard more intelligibly in far-end noisy environments. To obtain an effective reinforcement technique, we adopt the soft decision scheme incorporating a speech absence probability (SAP) in the frequency dependent signal-to-noise ratio (SNR) recovery method where the clean speech spectrum is estimated and the reinforcement gain is inherently derived and modified within the unified framework. Performance of the proposed method is evaluated by a subjective testing under various noisy environments. This is an improvement over previous approaches.

Improved A Posteriori Speech Presence Probability Estimation Based on a Likelihood Ratio With Fixed Priors

In this paper, we present an improved estimator for the speech presence probability at each time-frequency point in the short-time Fourier transform domain. In contrast to existing approaches, this estimator does not rely on an adaptively estimated and thus signal-dependent a priori signal-to-noise ratio estimate. It therefore decouples the estimation of the speech presence probability from the estimation of the clean speech spectral coefficients in a speech enhancement task. Using both a fixed a priori signal-to-noise ratio and a fixed prior probability of speech presence, the proposed a posteriori speech presence probability estimator achieves probabilities close to zero for speech absence and probabilities close to one for speech presence. While state-of-the-art speech presence probability estimators use adaptive prior probabilities and signal-to-noise ratio estimates, we argue that these quantities should reflect true a priori information that shall not depend on the observed signal. We present a detection theoretic framework for determining the fixed a priori signal-to-noise ratio. The proposed estimator is conceptually simple and yields a better tradeoff between speech distortion and noise leakage than state-of-the-art estimators.

Environment-Optimized Speech Enhancement

In this paper, we present a training-based approach to speech enhancement that exploits the spectral statistical characteristics of clean speech and noise in a specific environment. In contrast to many state-of-the-art approaches, we do not model the probability density function (pdf) of the clean speech and the noise spectra. Instead, subband-individual weighting rules for noisy speech spectral amplitudes are separately trained for speech presence and speech absence from noise recordings in the environment of interest. Weighting rules for a variety of cost functions are given; they are parameterized and stored as a table look-up. The speech enhancement system simply works by computing the weighting rules from the table look-up indexed by the a posteriori signal-to-noise ratio (SNR) and the a priori SNR for each subband computed on a Bark scale. Optimized for an automotive environment, our approach outperforms known-environment-independent-speech enhancement techniques, namely the a priori SNR-driven Wiener filter and the minimum mean square error (MMSE) log-spectral amplitude estimator, both in terms of speech distortion and noise attenuation.

A Probabilistic Combination Method of Minimum Statistics and Soft Decision for Robust Noise Power Estimation in Speech Enhancement

In this letter, we propose a novel approach to noise power estimation for robust speech enhancement in noisy environments. From investigation of the state-of-the-art techniques for noise power estimation, it is discovered that the previously known methods are accurate mostly either during speech absence or speech presence, but none of it works well in both situations. Our approach combines minimum statistics (MS) and soft decision (SD) techniques based on probability of speech absence. The performance of the proposed approach is evaluated by a quantitative comparison method and subjective test under various noise environments and found to yield better results compared with conventional MS- and SD-based schemes.

Complex laplacian probability density function for noisy speech enhancement

In this letter, we propose a new approach to speech enhancement based on a complex Laplacian probability density function (pdf). With a use of a goodness-of-fit (GOF) test, we discover that the complex Laplacian pdf is more desirable to describe noisy speech distribution than the conventional Gaussian pdf for speech enhancement. The likelihood ratio (LR) is computed and then applied to computation of the speech absence probability in the speech enhancement algorithm. Our evaluation shows that the proposed complex Laplacian pdf scheme provides further speech quality enhancement compared with the conventional method through a number of objective tests.

A new speech enhancement method for two-input two-output hearing aids

Human beings have the ability to pick up a speech signal in noisy environments, which is known as the cocktail-party effect. This phenomenon is, however, often degraded in persons with impaired hearing. Therefore, a good method to enhance this phenomenon must be very useful to improve speech intelligibility in ambient noise for hearing-impaired persons. In this lecture, multi-input two-output speech enhancement techniques are first summarized. Subsequently, a new two-input two-output speech enhancement method is proposed that is based on the frequency domain binaural model (FDBM) [Nakashima et al., Acoust. Sci. Technol. 24, 172–178 (2003)]. In the proposed method, the interaural differences are first calculated from the noisy observations and employed to determine the sound-source locations. A speech absence probability (SAP) estimator is then developed using knowledge of estimated source locations. It is further integrated into the original FDBM to improve the interference-suppression capability. Effectiveness and superiority of the proposed speech enhancement method with respect to the original FDBM are validated through comprehensive experiments in terms of objective and subjective speech-quality measures. [Work supported by Sendai Intelligent Knowledge Cluster.]

Multiple statistical models for soft decision in noisy speech enhancement

Most speech enhancement algorithms are based on the assumption that speech and noise are both Gaussian in the discrete cosine transform (DCT) domain. For further enhancement of noisy speech in the DCT domain, we consider multiple statistical distributions (i.e., Gaussian, Laplacian and Gamma) as a set of candidates to model the noise and speech. We first use the goodness-of-fit (GOF) test in order to measure how far the assumed model deviate from the actual distribution for each DCT component of noisy speech. Our evaluations illustrate that the best candidate is assigned to each frequency bin depending on the Signal-to-Noise-Ratio (SNR) and the Power Spectral Flatness Measure (PSFM). In particular, since the PSFM exhibits a strong relation with the best statistical fit we employ a simple recursive estimation of the PSFM in the model selection. The proposed speech enhancement algorithm employs a soft estimate of the speech absence probability (SAP) separately for each frequency bin according to the selected distribution. Both objective and subjective tests are performed for the evaluation of the proposed algorithms on a large speech database, for various SNR values and types of background noise. Our evaluations show that the proposed soft decision scheme based on multiple statistical modeling or the PSFM provides further speech quality enhancement compared with recent methods through a number of subjective and objective tests.