Year: 2024
Type: article
Abstract: This article develops mathematical methods of learning and decision-making in speech recognition, when extracting informative properties from speech signals, intelligent algorithms based on a data set consisting of properties. Effective ways to isolate features from speech signals are explained in sequential steps. These sequences are reflected in the database of several solutions (HMM, CNN and DTV) based on their comparative analysis, and the results are presented in the form of a table. In the experiments, sounds, syllables and words of speech were evaluated based on familiar intelligent algorithms, and a high result was achieved using the CNN mode.
Source: THE JOURNAL OF CIEES
Institution: Tashkent University of Information Technology
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4405660741