This paper analyzes the effectiveness of modern clustering algorithms for automatic segmentation of speech signals. Segmentation methods based on K-means, fuzzy c-means, and DBSCAN algorithms were applied, and the time boundaries of words were determined using these methods. The obtained results confirm the effectiveness of these algorithms in automatic speech processing. The work also describes an approach to defining precise word boundaries. The results were compared with other methods of speech signal segmentation.
| Mualliflar | Urinboev, Johongir, Nugmanova, Mavluda |
|---|---|
| Jurnal | Al-Farg'oniy avlodlari |
| Nashr sanasi | 2024-12-28 |
| Son | 4 |
| Betlar | 220-225 |
| Til | Rus |
Aвтоматическое распознавание речи, сегментация речевого сигнала, кластеризация, метод k-means, метод нечеткого c-means, метод DBSCAN, Biharmonic equation, semicircle, non-correct problem, approximate solution, Laplace operator, conditional correctness, stability theorem, Fourier series, regularization method, Hilbert space, Fredholm equation., Nutqni avtomatik tanib olish, nutq signalini segmentatsiyalash, klasterlash, k-means usuli, noaniq c-means usuli, DBSCAN usuli.
In this article, we have simulated the generation of a solar panel using a control power supply block in Simulink software. To generate the required power and voltage, solar panels are connected in series. We used a…
One of the important characteristics that determine the quality of the finished product is moisture. The most promising direction is optical hygrometry, the basis of which is the property of water to absorb infrared…
The features of implementing Massive MIMO technology in 5th generation mobile networks are analyzed and its quantitative and qualitative indicators are determined. The article also discusses the advantages and…
Fuzzy logic has become increasingly prominent in addressing uncertainties inherent in decision-making processes across various domains. A pivotal component of fuzzy systems is the rule base, which establishes the…
This article discusses the issue of analyzing sales data in the field of e-commerce using a desktop application. The application downloads sales data from a CSV file and calculates sales volumes by products, revenue by…
With advancements in satellite sensors, multi-source observation systems have become widely utilized. However, significant disparities exist among quantitative remote sensing products due to variations in observational…
Abstract. This article presents the results of a study of limiting and internal normal stresses depending on the moisture content and temperature of fruits (apricots and plums), vegetables (potatoes and carrots) during…
The article discusses the integration of hybrid methods of system analysis to improve the efficiency of decision-making under uncertainty and multiparameter constraints. Modern approaches to solving complex control and…
The work deals with the development of models and algorithms for intellectual analysis of data in the analysis and classification of textual data. With the help of the developed methods and models, it is used to solve…
Buildings consume a lot of energy, and inefficient lighting increases costs. The study presents a lighting control system based on Internet of Things technologies with PIR and Doppler sensors, an external light sensor…