Non-stationary Noise Types Research Articles

A context-based model to predict the intelligibility of sentences in non-stationary noises.

The context-based Extended Speech Transmission Index (cESTI) (van Schoonhoven et al., 2022, J. Acoust. Soc. Am. 151, 1404-1415) was successfully applied to predict the intelligibility of monosyllabic words with different degrees of context in interrupted noise. The current study aimed to use the same model for the prediction of sentence intelligibility in different types of non-stationary noise. The necessary context factors and transfer functions were based on values found in existing literature. The cESTI performed similar to or better than the original ESTI when noise had speech-like characteristics. We hypothesize that the remaining inaccuracies in model predictions can be attributed to the limits of the modelling approach with regard to mechanisms, such as modulation masking and informational masking.

Combined i-Vector and Extreme Learning Machine Approach for Robust Speaker Identification and Evaluation with SITW 2016, NIST 2008, TIMIT Databases

In this article, a novel combined i-vector and an Extreme Learning Machine (ELM) is proposed for speaker identification. The ELM is chosen because it is fast to train and has a universal approximator property. Four combinations of features based on Mel Frequency Cepstral Coefficient and Power Normalized Cepstral Coefficient are used. Besides, seven fusion methods are exploited. The system is evaluated with three different databases, namely: the SITW 2006, NIST 2008, and the TIMIT database. This work employs the 2016 SITW database for the first time for speaker identification using the integration between the ELM and i-vector approach. From each database, 120 speakers with 1200 speech utterances are used (overall 360 speakers with 3600 speech utterances). Furthermore, comprehensive evaluations are exploited with a wide range of realistic background noise types (Stationary noise AWGN and Non-Stationary Noise types) with the handset effect. The proposed system is compared with the Gaussian Mixture Model-Universal Background Model (GMM-UBM) and other states of the art approaches. The results show that the i-vector method outperforms the GMM-UBM approach and other state- of-the-art methods under specific conditions, and that fusion techniques can be used to improve robustness to noise and handset effects.

Speech enhancement by LSTM-based noise suppression followed by CNN-based speech restoration

Single-channel speech enhancement in highly non-stationary noise conditions is a very challenging task, especially when interfering speech is included in the noise. Deep learning-based approaches have notably improved the performance of speech enhancement algorithms under such conditions, but still introduce speech distortions if strong noise suppression shall be achieved. We propose to address this problem by using a two-stage approach, first performing noise suppression and subsequently restoring natural sounding speech, using specifically chosen neural network topologies and loss functions for each task. A mask-based long short-term memory (LSTM) network is employed for noise suppression and speech restoration is performed via spectral mapping with a convolutional encoder-decoder network (CED). The proposed method improves speech quality (PESQ) over state-of-the-art single-stage methods by about 0.1 points for unseen highly non-stationary noise types including interfering speech. Furthermore, it is able to increase intelligibility in low-SNR conditions and consistently outperforms all reference methods.

Thorough evaluation of TIMIT database speaker identification performance under noise with and without the G.712 type handset

In this work, a speaker identification system is proposed which employs two feature extraction models, namely: the power normalized cepstral coefficients and the mel frequency cepstral coefficients. Both features are subjected to acoustic modeling using a Gaussian mixture model–universal background model. The purpose of this work is to provide a thorough evaluation of the effect of different types of noise on the speaker identification accuracy (SIA) and thereby providing benchmark figures for future comparative studies. In particular, the additive white Gaussian noise and eight non-stationary noise types (with and without the G.712 type handset) corresponding to various signal to noise ratios are tested. Fusion strategies are also employed using late fusion methods: maximum, weighted sum, and mean fusion. The measurements of randomly selected 120 speakers from the TIMIT database are employed and the SIA is used to measure the system performance. The weighted sum fusion resulted in the best performance in terms of SIA with noisy speech. The proposed model given in this work and its related analysis paves the way for further work in this important area.

Evaluation of a speaker identification system with and without fusion using three databases in the presence of noise and handset effects

In this study, a speaker identification system is considered consisting of a feature extraction stage which utilizes both power normalized cepstral coefficients (PNCCs) and Mel frequency cepstral coefficients (MFCC). Normalization is applied by employing cepstral mean and variance normalization (CMVN) and feature warping (FW), together with acoustic modeling using a Gaussian mixture model-universal background model (GMM-UBM). The main contributions are comprehensive evaluations of the effect of both additive white Gaussian noise (AWGN) and non-stationary noise (NSN) (with and without a G.712 type handset) upon identification performance. In particular, three NSN types with varying signal to noise ratios (SNRs) were tested corresponding to street traffic, a bus interior, and a crowded talking environment. The performance evaluation also considered the effect of late fusion techniques based on score fusion, namely, mean, maximum, and linear weighted sum fusion. The databases employed were TIMIT, SITW, and NIST 2008; and 120 speakers were selected from each database to yield 3600 speech utterances. As recommendations from the study, mean fusion is found to yield overall best performance in terms of speaker identification accuracy (SIA) with noisy speech, whereas linear weighted sum fusion is overall best for original database recordings.

Analysis of the SNR Estimator for Speech Enhancement Using a Cascaded Linear Model

Elimination of tainted noise and improving the overall quality of a speech signal is speech enhancement. To gain the advantage of individual algorithms we propose a new linear model and that is in the form of cascade adaptive filters for suppression of non-stationary noise. We have successfully deployed NLMS (Normalized Least Mean Square) algorithm, Sign LMS (Least Mean Square) and RLS (Recursive Least Square) as the main de-noising algorithms. Moreover, we are successful in demonstrating that the prior information about the noise is not required otherwise it would have been difficult to estimate for fast-varying noise in non-stationary environment. This approach estimates clean speech by recognizing the long segments of the clean speech as one whole unit. During experiment/implementation we used in-house database (includes various types of non stationary noise) for speech enhancement and proposed model results have shown improvement over conventional algorithms not only in objective but in subjective evaluations as well. Simulations present good results with a new linear model that are compared with individual algorithm results.

A Corpus-Based Approach to Speech Enhancement From Nonstationary Noise

Temporal dynamics and speaker characteristics are two important features of speech that distinguish speech from noise. In this paper, we propose a method to maximally extract these two features of speech for speech enhancement. We demonstrate that this can reduce the requirement for prior information about the noise, which can be difficult to estimate for fast-varying noise. Given noisy speech, the new approach estimates clean speech by recognizing long segments of the clean speech as whole units. In the recognition, clean speech sentences, taken from a speech corpus, are used as examples. Matching segments are identified between the noisy sentence and the corpus sentences. The estimate is formed by using the longest matching segments found in the corpus sentences. Longer speech segments as whole units contain more distinct dynamics and richer speaker characteristics, and can be identified more accurately from noise than shorter speech segments. Therefore, estimation based on the longest recognized segments increases the noise immunity and hence the estimation accuracy. The new approach consists of a statistical model to represent up to sentence-long temporal dynamics in the corpus speech, and an algorithm to identify the longest matching segments between the noisy sentence and the corpus sentences. The algorithm is made more robust to noise uncertainty by introducing missing-feature based noise compensation into the corpus sentences. Experiments have been conducted on the TIMIT database for speech enhancement from various types of nonstationary noise including song, music, and crosstalk speech. The new approach has shown improved performance over conventional enhancement algorithms in both objective and subjective evaluations.

On-line Stochastic Matching compensation for non-stationary noise

This paper treats the problem of noise compensation in speech recognition when training and testing conditions do not match. We are interested in two types of non-stationary noise that may be present during test, namely slowly varying and abruptly varying noises. The context of our work is the Stochastic Matching framework. The Stochastic Matching compensation method transforms test data using an affine compensation function whose parameters are computed off-line. Stochastic Matching approaches are interesting since they make little assumptions about the nature and the level of the noise but they are best suited for the compensation of stationary noise. In this paper we propose an original contribution to the Stochastic Matching framework. It is based on an on-line frame-synchronous version of Stochastic Matching method to compensate for slowly varying noise. Our contribution extends this compensation algorithm in order to compensate for abruptly varying noise. The basic idea of the proposed methods is to perform the compensation and the recognition steps at the same time. The environment changes are identified using monitoring algorithms. The performance of our proposed methods is evaluated on two speech databases, one recorded in moving cars (VODIS), and another one obtained by corrupting VODIS with abruptly varying noise from NOISEX. The proposed approaches significantly outperform classical compensation methods (Off-line Stochastic Matching, Sequential Mean Cepstrum Removal, Parallel Model Combination, Spectral Subtraction). For instance, we obtain up to 32.6% word error rate reduction over S-MCR on database corrupted by a 10 dB abruptly varying white noise.

Noise processing techniques for time‐domain EM systems

A variety of signal processing techniques can be used to minimize the effects of noise on linear, wideband, electromagnetic (EM) systems operating in the time‐domain. All systems use repetitive waveforms with polarity reversal in alternate half‐cycles. Exponential averaging or digital integration (stacking) is employed to increase signal‐to‐noise (S/N) ratios by limiting the noise acceptance to narrow frequency bands centered on odd harmonics of the repetition frequency, the width of the acceptance bands being inversely proportional to stacking time. For certain types of nonstationary noise (e.g., occasional transients) or coherent noise (e.g., powerlines) it is possible to increase S/N ratios above those obtained by simple stacking for an equal time by use of techniques such as pruning, tapered stacking or randomized stacking. With some system waveforms and when the noise spectrum is not “white”, use of preemphasis filtering in the transmitter and a corresponding de‐emphasis filter in the receiver may significantly improve the input S/N ratio before stacking. Specific applications of the various techniques are discussed with reference to one particular time‐domain EM system, the UTEM 3 system. By their use, improvements in S/N ratio of as much as 6 to 1 have been regularly achieved without any increase in transmitter power, depending on the nature of the local noise.

On the Detection of Signals in Nonstationary Noise by Product Arrays

Two detection systems having multiple inputs each consisting of nonstationary noise and a stationary random signal are analyzed and compared on a signal-to-noise ratio basis. In both systems the inputs are divided into two groups and the sum of each group is formed. In the first system the resulting two wave forms are multiplied directly and the product averaged. In the second system the two wave forms are strongly clipped prior to multiplication, forming a polarity-coincidence correlator. Previous studies have shown the latter system to be slightly inferior for stationary noise. The results of this paper show that the latter system may be quite superior in certain types of nonstationary noise.