The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The results of noise-resistant speech signal processing, with the use of linear adaptive filters. To enhance the filtration efficiency, the filtration is proposed to use a cascade signal. This approach improves the noise resistance of the signal. A comparative analysis of the performance of the developed algorithm with known methods of filtration. The effectiveness of the proposed filter is demonstrated...
An automatic, text-independent speaker verification (SV) system is proposed using Line Spectral Frequency (LSF) features. The state-of-the-art Gaussian Mixture Model with Universal Background Model (GMM-UBM) framework is used for speaker modeling and verification. A score-level fusion based technique is employed in order to extract complementary information from static and dynamic LSF features and...
Information to be exchanged between two parties needs compression for achieving its efficient transmission. Encoded information gets distorted during its transmission over a channel due to noise. For monitoring and analysis of such noisy traffic of an adversary over communication networks, it is required to find the type of information, whether it is text or speech, then to restore it for further...
Objective measures are favored and widely used by many researchers in evaluating the quality of noise-suppressed speech. A good and reliable objective measure should have property that it could evaluate speech quality in consistent and well correlated with subjective ratings. In this paper, several widely used objective measures are applied to the speech signals with the Chinese languages including...
Speech enhancement and speech separation are important frontends of many speech processing systems. In real tasks, the background noises are often mixed with some human voice interferences. In this paper, we explore a framework to unify speech enhancement and speech separation for a speaker-dependent scenario based on deep neural networks (DNNs). Using a supervised method, DNN is adopted to directly...
Non-negative spectrogram decomposition and its variants have been extensively investigated for speech enhancement due to their efficiency in extracting perceptually meaningful components from mixtures. Usually, these approaches are implemented on the condition that training samples for one or more sources are available beforehand. However, in many real-world scenarios, it is always impossible for...
Objective evaluation of speech quality is an important part of the quality of communications service, and the speech in high-noise environments affect the people's auditory perception, which is an important factor for people to determine the speech quality. In this paper, a non-intrusive evaluation method for speech quality in high-noise environments is proposed, and a noise tracking and subtraction...
The speaker verification (SV) task has been an active area of research in the last thirty years. One of the recent research topics is on improving the robustness of SV system in challenging environments. This paper examines the robustness of current state of the art SV system against background noise corruptions. Specifically, we consider the scenario where the SV system is trained from noise free...
Dysphonia is a qualitative and quantitative alteration of the voice due to a structural or functional modification of one or more organs involved in voice production. Voice disorders are prevalent in certain working categories, particularly those of teachers, singers and actors. It is possible to evaluate the state of health of a voice through the acoustic analysis of the speech signal. This provides...
Reverberated speech signals in noisy acoustical environments cause some problems such as reducing speech intelligibility, distinguishing speakers, locating source, quality for hands-free telephony, hearing aid, etc. Adaptive filters can be applied to suppress the interfering signals and reduce the reverberation effects or to dereverberate the received speech signals at microphone. In this paper, Bayesian...
In this paper, we develop a Bayesian short-time spectral amplitude (STSA) estimator with the purpose of singlechannel speech enhancement in the presence of moderate levels of non-stationary noise. In this regard, we first apply a minimum mean squared error (MMSE) approach for the joint estimation of the short-term predictor (STP) parameters of the speech and noise signals, from the noisy speech observations...
The human ear has a great ability to isolate speech in a noisy environment and, therefore, constitutes a great source of inspiration for speech enhancement algorithms. In this work, we propose a Bayesian estimator for speech enhancement that integrates the cochlea's compressive nonlinearity in its cost function. When compared to existing Bayesian speech enhancement estimators, the proposed estimator...
This paper focuses on the development of an automatic sound classifier for digital hearing aids that aims to enhance the listening comprehension when the user goes from a sound environment to another different one. The approach consists in dividing the classifying algorithm into two layers that make use of two neural network algorithms that work more efficiently: the input signal discriminated by...
A method is presented for restoration of noisy bandlimited archived speech records. Speech is modeled with a formant-tracking linear prediction (FTLP) model of the spectral envelope and a harmonic noise model (HNM) of the excitation. The time-varying trajectories of the parameters of the LP and HNM models are tracked with Viterbi classifiers and denoised with Kalman filters. A frequency domain pitch...
Voiced sounds are the result of a periodic excitation of the vocal tract due to the vocal folds vibration. They are characterised by the fundamental frequency of the phonation, named pitch, which is the rate of vibration of the folds. In pathologic voices, pitch variations are indicative of the patient status, hence robust pitch estimation methods are required in order to track such variations and...
This paper addresses the problem of robust text-independent speaker verification when some of the features for the target signal are heavily masked by noise. In the framework of Gaussian mixture models (GMMs), a new approach based on the spectral subtraction technique and the statistical missing feature compensation is presented. The identity of spectral features missing due to noise masking is provided...
In this paper a partial knowledge of measurement noise is incorporated into the multidelay frequency domain adaptive filtering scheme to improve its performance in colored measurement noise scenarios. The proposed algorithm is obtained by minimizing a BLUE criterion function using the stochastic gradient method and then switching over to the frequency domain to reduce the computational complexity...
Most recognition methods, which have shown to be highly efficient under noise-free conditions fail dramatically with S/N ratios around or below 10 dB. One of the consequences of these high noise levels is that most Begin-End Point Detectors fail to separate properly the speech segments of the noise ones. Therefore, the speech recognition mechanisms will not have a clear boundary to start the processing...
In this paper we focus on the problem of acoustic echo cancellation and noise reduction for hands-free telephony devices. A standard echo canceller is combined with a frequency domain post-filter, which applies a novel psychoacoustically motivated weighting rule. The algorithm makes use of the masking threshold of the human auditory system to achieve a perceived reduction of noise and residual echo...
This paper deals with the enhancement of speech corrupted by real additive noises in a car when two observations are available. As far as we know, no enhancement system was capable of improving both the quality and the intelligibility of the noisy signals. We propose an enhancement method using thresholding, segmentation and filtering in subband domain. The main idea is to expand in subband signals...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.