In traditional speech processing, feature extraction and classification were conducted as separate steps. The advent of deep neural networks has enabled methods that simultaneously model the relationship between acoustic and phonetic characteristics of speech while classifying it directly from the raw waveform. The first convolutional layer in these networks acts as a filter bank. To enhance interpretability and reduce the number of parameters, researchers have explored the use of parametric filters, with the SincNet architecture being a notable advancement. In SincNet's initial convolutional layer, rectangular bandpass filters are learned instead of fully trainable filters. This approach allows for modeling with fewer parameters, thereby improving the network's convergence speed and accuracy. Analyzing the learned filter bank also provides valuable insights into the model's performance. The reduction in parameters, along with increased accuracy and interpretability, has led to the adoption of various parametric filters and deep architectures across diverse speech processing applications. This paper introduces different types of parametric filters and discusses their integration into various deep architectures. Additionally, it examines the specific applications in speech processing where these filters have proven effective.

In this contribution, a modified gray wolf Algorithm for use in engineering applications is presented. The grey wolf Algorithm is one of the meta-heuristic optimization methods that has recently been widely used by researchers due to its good capabilities. The mechanism is free of derivation, simple in execution and implementation, and only needs target function as input of the problem, among other things that make the gray wolf Algorithm popular and of interest. But the problem that can be mentioned about it is that the decreasing factor used in it is linear and in some non-linear problems, it may cause more error or late convergence to the original solution. This bottleneck is solved by presenting a modified grey wolf Algorithm. Then the results are compared in the form of an applied numerical example in engineering sciences with the classic grey wolf Algorithm and some similar proposed coefficients to determine the efficiency of the modified Algorithm.

In today's knowledge-oriented world, there are many problems that need to be solved in a structured and systematic way. In this regard, an effective solution is to use suitable Algorithmic approaches and perspectives. Algorithm includes a set of processes and instructions that can be used to solve many computer problems or real world problems in a regular and effective way. The use of Algorithms in solving systematic problems saves time and money and provides new and valuable solutions. In fact, the Algorithm is considered the beating heart of potentially powerful technologies such as artificial intelligence and today, instead of thinking about the future of Algorithms, people consider the future to belong to Algorithms. Although Algorithms have relatively the same concept in different fields of implementation, but in order to achieve specific applications of the desired Algorithm in different fields, we must take advantage of the approaches and perspectives of appropriate Algorithms. Therefore, according to the importance of the subject, in this article, we intend to investigate the different dimensions of Algorithmic perspectives and describe the common operational and practical classifications of important Algorithms in computer science and related technology-oriented fields.

The split common fixed point problem in Banach spaces, is to find a point which belongs to the set of  common fixed points of a family of nonlinear  mappings  in a  Banach space such that its image under a linear transformation belongs to the set of  common fixed point of another family of nonlinear mappings in a Banach space  in the image space.  The split common fixed point problem have  received much attention due to its applications in  many areas of mathematics and engineering sciences such as  optimization,  image reconstruction,  intensity-modulated radiation therapy and  image  processing. In this  paper we present a new Algorithm for solving the  split common fixed point problem for a finite family of  generalized demimetric mappings in uniformly convex and smooth Banach spaces.We establish  strong convergence of the sequence  generated by the  Algorithm  to a solution of split common fixed point problem  which also solves some variational inequality problems. Finally we present some applications of our main result for solving the multiple-set split feasibility problem and the split null point problem.