Modified coherence‐based dictionary learning method for speech enhancement

In this paper, we propose a novel speech enhancement (SE) method by exploiting the discrete wavelet transform (DWT). This new method reduces the amount of fast time-varying portion, viz. the DWT-wise detail component, in the spectrogram of speech signals so as to highlight the speech-dominant component and achieves better speech quality. A particularity of this new method is that it is completely unsupervised and requires no prior information about the clean speech and noise in the processed utterance. The presented DWT-based SE method with various scaling factors for the detail part is evaluated with a subset of Aurora-2 database, and the PESQ metric is used to indicate the quality of processed speech signals. The preliminary results show that the processed speech signals reveal a higher PESQ score in comparison with the original counterparts. Furthermore, we show that this method can still enhance the signal by totally discarding the detail part (setting the respective scaling factor to zero), revealing that the spectrogram can be down-sampled and thus compressed without the cost of lowered quality. In addition, we integrate this new method with conventional speech enhancement algorithms, including spectral subtraction, Wiener filtering, and spectral MMSE estimation, and show that the resulting integration behaves better than the respective component method. As a result, this new method is quite effective in improving the speech quality and well additive to the other SE methods.

This work presents the use of Dictionary Learning methods for a sparse representation of 4D seismic data in history matching. We consider a trade-off between the number of coefficients retained in the sparse data representation, the computational cost, and how well we can capture the main features of the 4D seismic signal. K-SVD is an iterative algorithm used in Dictionary Learning problems that alternates between the calculation of the sparse representation vector and dictionary update. For the definition of the sparse representation vector, one can constrain the problem into two distinct approaches: (1) sparsity-constrained; and (2) error-constrained. We evaluated the two methods and the influence of critical parameters of the algorithm (dictionary size, number of iterations, patch size, and training dataset size) using a synthetic reservoir model. Results showed that regardless of which of the constrained approaches we used, the dictionary learning method can capture the main features of the 4D seismic signal with a sparse representation. However, the number of nonzero coefficients is highly dependent on the approach. Besides, dictionary size, number of iterations, patch size, and training dataset size, also have a significant impact on the number of coefficients and the computational cost. Consequently, the selection of these variables is crucial as it might affect the history-matching process. In a permanent-reservoir-monitoring (big data) scenario, the use of sparse representation allows us to retain the main features of the reservoir in a reasonably sized data representation that can be directly used in a history matching method, leading to improved reservoir characterization and a better understanding of the reservoir properties. This paper gives a practical implementation of a technique already used in other areas (image denoising, processing of raw seismic data, and facies representation) applied to 4D seismic data.

Speech enhancement is a process which cultivates the quality of speech signal in noisy environment. It refers to removing or reducing the background noise in order to obtain an improved quality of original speech signal. Degradation of speech signal is most common problem in speech communication, so enhancement of the speech plays a vital role in improving the quality of speech signal. A number of methods are used for speech enhancement. Here we are using LPC based FEM in Wiener filter for speech enhancement. This method is then compared with several speech enhancement algorithms to obtain a better result or better speech quality. On comparing this method with other speech enhancement methods we obtain better speech performance. Here we are using NOIZEUS speech database in order to compare different speech enhancement methods. The experimental result shows that our proposed method provides better result and there is no information loss in original speech signal.

Speech enhancement (SE) methods mainly focus on recovering clean speech from noisy input. In real-world speech communication, however, noises often exist in not only speaker but also listener environments. Although SE methods can suppress the noise contained in the speaker’s voice, they cannot deal with the noise that is physically present in the listener side. To address such a complicated but common scenario, we investigate a deep learning-based joint framework integrating noise reduction (NR) with listening enhancement (LE), in which the NR module first suppresses noise and the LE module then modifies the denoised speech, i.e., the output of the NR module, to further improve speech intelligibility. The enhanced speech can thus be less noisy and more intelligible for listeners. Experimental results show that our proposed method achieves promising results and significantly outperforms the disjoint processing methods in terms of various speech evaluation metrics.

In real-world scenarios, speech signals are inevitably corrupted by various types of interference, making speech enhancement (SE) a critical task for robust speech processing. However, most existing SE methods only handle a limited range of distortions, such as additive noise, reverberation, or band limitation, while the study of SE under multiple simultaneous distortions remains limited. This gap affects the generalization and practical usability of SE methods in real-world environments. To address this gap, this paper proposes a novel Universal Discrete-domain SE model called UDSE. UDSE primarily works to enhance speech by predicting clean discrete tokens that are quantized by the residual vector quantizer (RVQ) of a pre-trained neural speech codec and are predicted in accordance with the RVQ rules. Specifically, UDSE first extracts global features from the degraded speech. Guided by these global features, the clean token prediction for each VQ follows the rules of RVQ, where the prediction of each VQ relies on the results of the preceding ones. Finally, the predicted clean tokens from all VQs are decoded to reconstruct the clean speech waveform. During training, the UDSE model employs a teacher-forcing strategy, and is optimized with cross-entropy loss. Experimental results confirm that the proposed UDSE model can effectively enhance speech degraded by various conventional and unconventional distortions, e.g., additive noise, reverberation, band limitation, clipping, phase distortion, and compression distortion, as well as their combinations. These results demonstrate the superior universality and practicality of UDSE compared to advanced regression-based SE methods.

Supervised models for speech enhancement are trained using artificially generated mixtures of clean speech and noise signals. However, the synthetic training conditions may not accurately reflect real-world conditions encountered during testing. This discrepancy can result in poor performance when the test domain significantly differs from the synthetic training domain. To tackle this issue, the UDASE task of the 7th CHiME challenge aimed to leverage real-world noisy speech recordings from the test domain for unsupervised domain adaptation of speech enhancement models. Specifically, this test domain corresponds to the CHiME-5 dataset, characterized by real multi-speaker and conversational speech recordings made in noisy and reverberant domestic environments, for which ground-truth clean speech signals are not available. In this paper, we present the objective and subjective evaluations of the systems that were submitted to the CHiME-7 UDASE task, and we provide an analysis of the results. This analysis reveals a limited correlation between subjective ratings and several supervised nonintrusive performance metrics recently proposed for speech enhancement. Conversely, the results suggest that more traditional intrusive objective metrics can be used for in-domain performance evaluation using the reverberant LibriCHiME-5 dataset developed for the challenge. The subjective evaluation indicates that all systems successfully reduced the background noise, but always at the expense of increased distortion. Out of the four speech enhancement methods evaluated subjectively, only one demonstrated an improvement in overall quality compared to the unprocessed noisy speech, highlighting the difficulty of the task. The tools and audio material created for the CHiME-7 UDASE task are shared with the community Key Terms: Speech Enhancement, CHIME-7 Challenge ,Deep Neural Networks, UDASE Task, Audio Restoration, Subjective Evaluation, objective Evaluation

Most algorithms for speech enhancement in the spectral domain focus on the acquisition of an estimator of the clean speech parameter, such as spectrum or amplitude. Enhanced speech quality with these methods mainly depends upon the accuracy of the estimator. When the signal-to-noise ratio (SNR) becomes lower, e.g., less than 3 dB, the enhanced speech often shows unsatisfactory quality. Here, we propose a new method for speech enhancement in the spectral domain. It is based on our view that speech characteristics are perceived mainly by part of the spectral components which have higher local (or instantaneous) SNRs. These components have a special importance on speech SNR enhancement and speech understanding. Identifying these components in the noisy spectrum with a decision system and constructing enhanced speech just by the same noisy amplitude and phase, the result is similar to those popular algorithms. All that is needed is a local SNR decision for each spectral component. The threshold for the decision is derived and two popular methods for local SNR (LSNR) estimation are suggested. It shows better performance under lower SNR signals. Though the number of these kind of spectral components varies with different signal SNR, any algorithms which can pick up more than 90% of it is sufficient to match any existing methods. Primary considerations and results are shown.

Deep learning-based speech enhancement (SE) methods predominantly draw upon two architectural frameworks: generative adversarial networks and diffusion models. In the realm of SE, capturing the local and global relations between signal frames is crucial for the success of these methods. These frameworks typically employ a UNet architecture as their foundational backbone, integrating Long Short-Term Memory (LSTM) networks or attention mechanisms within the UNet to effectively model both local and global signal relations. However, the coupled relation modeling way may not fully harness the potential of these relations. In this paper, we propose an innovative Association-based Fusion Speech Enhancement method (AFSE), a decoupled method. AFSE first constructs a graph that encapsulates the association between each time window of the speech signal, and then models the global relations between frames by fusing the features of these time windows in a manner akin to graph neural networks. Furthermore, AFSE leverages a UNet with dilated convolutions to model the local relations, enabling the network to maintain a high-resolution representation while benefiting from a wider receptive field. Experimental results demonstrate that the AFSE method significantly improves performance in speech enhancement tasks, validating the effectiveness and superiority of our approach. The code is available at https://github.com/jie019/AFSE_IJCAI2025.

Automatic prostate segmentation from MR images is an important task in various clinical applications such as prostate cancer staging and MR-guided radiotherapy planning. However, the large appearance and shape variations of the prostate in MR images make the segmentation problem difficult to solve. Traditional Active Shape/Appearance Model (ASM/AAM) has limited accuracy on this problem, since its basic assumption, i.e., both shape and appearance of the targeted organ follow Gaussian distributions, is invalid in prostate MR images. To this end, the authors propose a sparse dictionary learning method to model the image appearance in a nonparametric fashion and further integrate the appearance model into a deformable segmentation framework for prostate MR segmentation. To drive the deformable model for prostate segmentation, the authors propose nonparametric appearance and shape models. The nonparametric appearance model is based on a novel dictionary learning method, namely distributed discriminative dictionary (DDD) learning, which is able to capture fine distinctions in image appearance. To increase the differential power of traditional dictionary-based classification methods, the authors' DDD learning approach takes three strategies. First, two dictionaries for prostate and nonprostate tissues are built, respectively, using the discriminative features obtained from minimum redundancy maximum relevance feature selection. Second, linear discriminant analysis is employed as a linear classifier to boost the optimal separation between prostate and nonprostate tissues, based on the representation residuals from sparse representation. Third, to enhance the robustness of the authors' classification method, multiple local dictionaries are learned for local regions along the prostate boundary (each with small appearance variations), instead of learning one global classifier for the entire prostate. These discriminative dictionaries are located on different patches of the prostate surface and trained to adaptively capture the appearance in different prostate zones, thus achieving better local tissue differentiation. For each local region, multiple classifiers are trained based on the randomly selected samples and finally assembled by a specific fusion method. In addition to this nonparametric appearance model, a prostate shape model is learned from the shape statistics using a novel approach, sparse shape composition, which can model nonGaussian distributions of shape variation and regularize the 3D mesh deformation by constraining it within the observed shape subspace. The proposed method has been evaluated on two datasets consisting of T2-weighted MR prostate images. For the first (internal) dataset, the classification effectiveness of the authors' improved dictionary learning has been validated by comparing it with three other variants of traditional dictionary learning methods. The experimental results show that the authors' method yields a Dice Ratio of 89.1% compared to the manual segmentation, which is more accurate than the three state-of-the-art MR prostate segmentation methods under comparison. For the second dataset, the MICCAI 2012 challenge dataset, the authors' proposed method yields a Dice Ratio of 87.4%, which also achieves better segmentation accuracy than other methods under comparison. A new magnetic resonance image prostate segmentation method is proposed based on the combination of deformable model and dictionary learning methods, which achieves more accurate segmentation performance on prostate T2 MR images.

Sparse and redundant representation of data assumes an ability to describe signals as linear combinations of a few atoms from a dictionary. If the model of the signal is unknown, the dictionary can be learned from a set of training signals. Like the K-SVD, many of the practical dictionary learning algorithms are composed of two main parts: sparse-coding and dictionary-update. This paper first proposes a Stagewise least angle regression (St-LARS) method for performing the sparse-coding operation. The St-LARS applies a hard-thresholding strategy into the original least angle regression (LARS) algorithm, which enables it to select many atoms at each iteration and thus results in fast solutions while still provides good results. Then, a dictionary update method named approximated singular value decomposition (ASVD) is used on the dictionary update stage. It is a quick approximation of the exact SVD computation and can reduce the complexity of it. Experiments on both synthetic data and 3-D image denoising demonstrate the advantages of the proposed algorithm over other dictionary learning methods not only in terms of better trained dictionary but also in terms of computation time.

Slip/antiblocking agents

Speech enhancement is an important pre-processing step in a wide diversity of practical fields related to speech signals, and many signal-enhancement methods, such as fixed beamforming and adaptive beamforming, have already been proposed for speech enhancement. However, lack of comprehensive and quantitative performance evaluation of current enhancement methods when multi-speeches exist simultaneously, it is difficult to choose an appropriate enhancement method in a practical application. This work aims to study the implementation of several enhancement methods for multi-speech enhancement in indoor environment of T60 = 0 s and T60 = 0.3 s. We first propose and compare two types of enhancement approaches. The first type is the basic enhancement methods, including the delay-and-sum beamforming (DSB), minimum variance distortionless response (MVDR), linearly constrained minimum variance (LCMV), and independent component analysis (ICA); The second type is the improved enhancement methods, including the robust MVDR and LCMV, realized by eigen decomposition and diagonal loading. In addition, the online enhancement performance based on the iteration of single-frame speech signals is then researched. The experimental results show that the enhancement effect of LCMV and ICA is relatively more stable in the case of the basic enhancement methods; in the case of the improved enhancement algorithms, the methods which employed diagonal loading iterations are better. In the case of the online enhancement, the DSB with frequency masking yields the best performance on the signal-to-interferences ratio (SIR) and possesses the capacity to suppress interferences.

The applications of speech fractal dimension to speech segmentation and enhancement are researched. The computer simulation results show that the new methods of speech segmentation and enhancement can achieve satisfactory performances.

In this paper, a speech enhancement method based on noise compensation performed on short time magnitude spectra is presented following the geometric approach. Here the noise estimate to be subtracted from the noisy speech spectrum is proposed to be determined by exploiting the low frequency regions of noisy speech of the current frame rather than depending only on the initial silence frames. We argue that this approach of noise estimation offers the capability of tracking the time variation of the non-stationary noise. A noise estimation confidence parameter is also introduced to prevent the subtraction of the overestimated noise. A procedure is thus formulated to compensate for the distortion in the magnitude spectrum of the noisy speech signal. The noise compensated magnitude spectrum is then recombined with the noisy unchanged phase spectrum to produce a modified complex spectrum prior to synthesize an enhanced frame. Extensive simulations are carried out using NOIZEUS database in order to evaluate the performance of proposed method. It is shown in terms of objective measures, spectrogram analysis and informal subjective listening test that the proposed method consistently outperforms one of the state-of-the-art methods of speech enhancement from noisy speech corrupted by babble or car noise at high as well as very low levels of SNR.

Sparse coding allows the representation of complex data as a linear combination of basis sparse vectors (alternatively called atoms or codewords), a collection of which constitutes a dictionary. Dictionary learning is a learning process aimed at finding a small number of optimal basis vectors for a more accurate representation of the original data. The existing dictionary learning methods do not address the inherent uncertainty of the input data in their learning processes. To compensate for the uncertainty, and to obtain a flexible and effective learning system, we introduce a new adaptive fuzzy dictionary learning (AFDL) method for image classification purposes. The new method iteratively alternates between sparse coding based on a given dictionary and an adaptive fuzzy dictionary learning approach to learn (improve) dictionary atoms. The adjustability of the dictionary and coefficients vectors, in this method, provide us a more accurate and straight representation of input data. AFDL was applied on magnetic resonance images from the cancer image archive datasets, for medical image classification of cancer tumors. Finally, the overall experimental results clearly show that our approach outperforms its rival techniques in terms of accuracy, sensitivity, and specificity. Convergence speed in the experimental results shows that AFDL can achieve its acceptable precision in a reasonable time.