CNN-QTLBO: an optimal blind source separation and blind dereverberation scheme using lightweight CNN-QTLBO and PCDP-LDA for speech mixtures

Sheeja, Jasmine J. C.; Sankaragomathi, B.

doi:10.1007/s11760-021-02084-0

Cited by 9 publications

(4 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sheeja et al proposed a lightweight convolutional neural network quantum teaching optimization algorithm based on discrete Fourier transform for MSS problems under noise interference. This algorithm effectively improved separation performance while eliminating blind separation noise [14].…”

Section: Introductionmentioning

confidence: 96%

Application of Unet-SE-Bisru Algorithm for Music Signal Processing in Music Source Separation

Zhang

2024

SCPE

View full text Add to dashboard Cite

At present, the use of time-domain deep learning for end-to-end neural network models has the problem of long training time and poor performance in music source separation. To address this issue, a U-network squeezing excitation bidirectional simple recursive unit model was proposed based on the deep extractor model. Replace Unet SE Bisru with Unet SE Bisru in the following text. This model improves the bidirectional long short-term memory network into a bidirectional simple recurrent unit, and then introduces attention mechanisms in the generalized encoding and decoding layers. The squeezing excitation block is used to selectively extract features based on the type of audio to be separated. Finally, group normalization is added after onedimensional convolution, And its effectiveness was verified. The experimental results show that the signal noise distortion ratio in the improved model is 5.68 decibels compared to the bidirectional simple recursive unit value, which is higher than the 5.55 decibels of bidirectional long short-term memory. After adding the squeezing excitation module, the overall increase is about 0.1-0.5 decibels. In addition, in the model comparison, the three indicators of the improved model with the same number of channels were 5.68 decibels, 5.91 decibels, and 11.28 decibels, respectively, higher than the benchmark model. Compared with other music source classification models, the improved model has better comprehensive separation performance. Although some indicators are lower than the comparison model, the signal noise distortion ratio of drum and bass is 6.11 decibels and 6.36 decibels, which is better than the comparison model. Overall, the improved model has high performance in music source separation for music signal processing and can be effectively applied in practical music source separation.

show abstract

Section: Introductionmentioning

confidence: 96%

Application of Unet-SE-Bisru Algorithm for Music Signal Processing in Music Source Separation

Zhang

2024

SCPE

View full text Add to dashboard Cite

show abstract

“…In such cases, the speeches from these devices are merged and transmitted to a designated receiver node. The primary objective is to differentiate and recover the individual speeches by leveraging the available perceptual data, i.e., BSS of audio files [13]. As evident from the observations, the development of a robust framework capable of effectively separating speech and music has the potential to yield substantial benefits across numerous lucrative applications [14], [15].…”

Section: Introductionmentioning

confidence: 99%

A Robust Hybrid Neural Network Architecture for Blind Source Separation of Speech Signals Exploiting Deep Learning

Ansari,

Alnajjar,

Khater

et al. 2023

IEEE Access

View full text Add to dashboard Cite

In the contemporary era, blind source separation has emerged as a highly appealing and significant research topic within the field of signal processing. The imperative for the integration of blind source separation techniques within the context of beyond fifth-generation and sixth-generation networks arises from the increasing demand for reliable and efficient communication systems that can effectively handle the challenges posed by high-density networks, dynamic interference environments, and the coexistence of diverse signal sources, thereby enabling enhanced signal extraction and separation for improved system performance. Particularly, audio processing presents a critical domain where the challenge lies in effectively handling files containing a mixture of human speech, silence, and music. Addressing this challenge, speech separation systems can be regarded as a specialized form of human speech recognition or audio signal classification systems that are leveraged to separate, identify, or delineate segments of audio signals encompassing human speech. In various applications such as volume reduction, quality enhancement, detection, and identification, the need arises to separate human speech by eliminating silence, music, or environmental noise from the audio signals. Consequently, the development of robust methods for accurate and efficient speech separation holds paramount importance in optimizing audio signal processing tasks. This study proposes a novel three-way neural network architecture that incorporates transfer learning, a pre-trained dual-path recurrent neural network, and a transformer. In addition to learning the time series associated with audio signals, this network possesses the unique capability of direct context-awareness for modeling the speech sequence within the transformer framework. A comprehensive array of simulations is meticulously conducted to evaluate the performance of the proposed model, which is benchmarked with seven prominent state-of-the-art deep learning-based architectures. The results obtained from these evaluations demonstrate notable advancements in multiple objective metrics. Specifically, our proposed solution showcases an average improvement of 4.60% in terms of short-time objective intelligibility, 14.84% in source-to-distortion ratio, and 9.87% in scale-invariant signal-to-noise ratio. These extraordinary advancements surpass those achieved by the nearest rival, namely the dual-path recurrent neural network time-domain audio separation network, firmly establishing the superiority of our proposed model's performance.

show abstract

“…But the major problem is that the useful information itself is unknown or rather the available information contains a set of overlapping useful information and they have to be extracted independently, and the problem becomes more complex when the information is affected by noise which degrades the intelligibility of the data set. In signal processing, one of the disciplines that deals with this problem of retrieving useful information is called blind source separation (BSS) [1][2][3][4]. The objective in solving the BSS problem is to recover a set of signals called source signals from a set of received signals called mixtures without prior knowledge either of the nature of the source signals or of the mechanism by which the mixing system (propagation medium) mixes the signals.…”

Section: Introductionmentioning

confidence: 99%

Source recovery by analytical maximization of phase-shifted kurtosis for the mixtures of noisy and noiseless signals

SMATTI

Arar

2023

Preprint

View full text Add to dashboard Cite

This manuscript presents a work that provides a study as well as a simple analytical solution for solving the blind source separation problem (BSS) for noiseless and noisy linear mixing of statistically independent stationary and nonstationary signals. The study is based on the exploitation of the probabilistic characteristics of the mixed signals by using the statistics of the second order and the fourth order for the completion of the separation. The proposed solution consists mainly of two steps based on the concept of the geometric solution. For the case of the mixture of two sources (2×2), the first step aims to transform the dependent signals into orthogonal signals (whitening) via the principal component analysis (PCA) principle. After the application of the PCA and in order to complete the statistical independence of the two uncorrelated signals, the second step aims to determine an adequate rotating angle that leads directly to the separation, and this angle is determined in this work analytically by the simple calculation of the phase shift of a sinusoidal objective function based on the sum of the kurtosis of the whitened signals. In the case of several sources (n×n), the solution (2×2) can be applied by a simple generalization which leads to the global separation. Whether for the noisy or noiseless case, the results obtained prove the reliability and efficiency by applying this analytical solution to achieve the desired objective, in particular by comparing the proposed algorithm with the application of two other separation algorithms, one of which involves the application of optimization techniques

show abstract

CNN-QTLBO: an optimal blind source separation and blind dereverberation scheme using lightweight CNN-QTLBO and PCDP-LDA for speech mixtures

Cited by 9 publications

References 25 publications

Application of Unet-SE-Bisru Algorithm for Music Signal Processing in Music Source Separation

Application of Unet-SE-Bisru Algorithm for Music Signal Processing in Music Source Separation

A Robust Hybrid Neural Network Architecture for Blind Source Separation of Speech Signals Exploiting Deep Learning

Source recovery by analytical maximization of phase-shifted kurtosis for the mixtures of noisy and noiseless signals

Contact Info

Product

Resources

About