This paper proposes a speech recognition system for automatically recognizing separately pronounced Uzbek words. The system uses MFCC for acoustic feature extraction and a Gaussian Hidden Markov Model for word modeling. The dataset was created from 15 participants, including 9 males and 6 females. Recordings were collected in a noise-free environment at 16,000 Hz with durations of 3–5 seconds. For experiments, 190 recordings were used for training and 20 for testing. The system achieved 84.2% accuracy, 87% precision, and 83% recall, demonstrating promising performance for Uzbek word recognition tasks overall
| Mualliflar | Nuritdinov, Nurbek |
|---|---|
| Jurnal | Al-Farg'oniy avlodlari |
| Nashr sanasi | 2026-03-18 |
| Son | 1 |
| Betlar | 270-275 |
| Til | Rus |
Speech signal, Mel-frequency cepstral coefficient (MFCC), Hidden Markov model (HMM), Data set, Segmentation and windowing, Fast Fourier transform (FFT)., речевой сигнал, мел-частотный кепстральный коэффициент (MFCC), скрытая марковская модель (HMM), набор данных, сегментация и оконная обработка, быстрое преобразование Фурье (FFT)., Nutq signali, Mel-chastotali kepstral koeffitsiyent (MFCC), Yashirin Markov modeli (HMM), Ma’lumotlar to‘plami, Segmentlash va oynalash, Tezkor Furye o‘zgartirishi (FFT).
This article analyzes two methods of adversarial attacks created against machine — training — based pest program detection systems-JSMA (Jacobian Saliency Map Attack) and Carlini & Wagner (C&W) - in a practical…
In this work, methods and algorithms for improving the quality of a given image are studied. Methods for reducing noise, improving contrast, increasing sharpness, and clearly displaying image elements during image…
This article presents a study of a dictionary-based pre-filter for safe educational content generation in a school environment. An Android application based on the MVVM architecture was developed as an experimental…
This article describes the research conducted by training the neural network model proposed by NVIDIA in speech recognition with a speech dataset of the Uzbek language. Currently, various models of neural networks are…
This article analyzes current trends in the field of cybersecurity. Today, the stable and protected operation of data transmission networks, computer systems, and mobile devices has become a fundamental condition…
While large-scale pre-trained models have significantly advanced multilingual Automatic Speech Recognition (ASR), many low-resource languages remain under-served due to the scarcity of high-quality annotated speech…
The scarcity of labeled data remains a fundamental limitation of supervised learning models. This study proposes an autoencoder-based unsupervised approach for detecting anomalies in network traffic. Experiments…
This article studies the effectiveness of modern deep learning models for the task of deep contextual analysis of text data in social networks. As part of the study, the ability of RNN, LSTM and DistilBERT models based…
This article analyzes the effectiveness of malware detection methods using artificial intelligence technologies. With the rapid development of modern information and communication technologies, the number of malicious…
This article studies the technical architecture and mathematical model of an adaptive digital environment based on the integration of AutoCAD + Python + AI in teaching engineering geometry and computer graphics. Within…