• Home
  • Search
  • Методи і технології біометричного розпізнавання суб’єкта за голосом
  • https://doi.org/10.33099/2304-2745/2018-2-63/110-113Copy DOI Icon

Методи і технології біометричного розпізнавання суб’єкта за голосом

  • Abstract
  • Literature Map
  • References
  • Similar Papers
Abstract

Speech recognition refers to biometric methods of personal identification along with fingerprint identification, hand configuration, shape and features of the face, retina, or gene structure. One of the main incentives for the study of speech recognition processes is the desire for more reliable recognition of the personality, since the uniqueness of the biometric characteristics provides a higher reliability of identification.The purpose of the article. Analysis of the difference between the verification and identification of the speech signal, which determines the different formulation of the problem and different methods for its solution. In addition, an analysis of the physical characteristics of the formation of the voice, determining the individual feature of the tonality of the subject.Subject recognition systems are divided into two types by voice: verification and speaker identification. During verification, the received voice sample is compared with the standard and its identity with that standard is established. It is necessary to make a decision out of two possible: there is a subject to those for whom it claims to be, or not.The task of identifying the subject is significantly different from the task of verification, it is necessary to identify from the set of N subjects who, according to their voice characteristics, will coincide with the characteristics of previously stored standards.The speech signal is a complex, multifaceted phenomenon, the parameters of which depend on a number of features of each person. The choice of the analyzed parameters, which can be used to recognize the subject, should be based on the objectives, methods and objectives of recognition. In modern voice recognition systems, people tend to use as much high-level information as possible, which is closely related to low-level information. The probability of correct recognition of the speaker increases in proportion to which signs are analyzed, the number of signs and their correlation.If during verification a solution can be formed using automatic systems, then today there are no clear characteristics when identifying by which it is possible to identify similarities between the disguised voice and the reference, since there are no algorithms for reducing the disguised voice to the “normal” pattern. The combination of subjective and objective methods can improve the accuracy of recognition, but the probability of correct identification of the subject in the case of masking the voice is still low.

Similar Papers
  • Book Chapter

An Overview of the Concept of Speaker Recognition

  • Dec 06, 2019
  • Intelligent Systems
  • Sindhu Rajendran +5
  • Research Article
  • Citations2

Учет многофакторности характеристик голоса в задачах идентификации диктора

  • Jul 05, 2021
  • International Scientific Technical Journal "Problems of Control and Informatics"
  • Виктор Иванович Соловьев +4
  • Book Chapter
  • Citations5

Automatic Speaker Recognition by Speech Signal

  • Oct 01, 2008
  • Milan Sigmund
  • Book Chapter
  • Citations1

Hybrid Feature Extraction Technique for Tamil Automatic Speech Recognition System in Noisy Environment

  • Jan 01, 2022
  • S Girirajan +1
  • Research Article
  • Citations15

A study on model-based error rate estimation for automatic speech recognition

  • Nov 01, 2003
  • IEEE Transactions on Speech and Audio Processing
  • Chao-Shih Huang +2
  • Conference Article
  • Citations3

Significance of Prosody Modification in Privacy Preservation on speaker verification

  • May 24, 2022
  • Control theory & applications
  • Ayush Agarwal +3
  • Research Article
  • Citations103

Fifty years of progress in speech and speaker recognition

  • Oct 01, 2004
  • The Journal of the Acoustical Society of America
  • Sadaoki Furui
  • Research Article
  • Citations25

Enhancements in automatic Kannada speech recognition system by background noise elimination and alternate acoustic modelling

  • Jan 22, 2020
  • International Journal of Speech Technology
  • G Thimmaraja Yadava +1
  • Conference Article
  • Citations9

Speech recognition of different sampling rates using fractal code descriptor

  • Jul 01, 2016
  • Rattaphon Hokking +2
  • PDF
  • Research Article
  • Citations12

Single-channel dereverberation by feature mapping using cascade neural networks for robust distant speaker identification and speech recognition

  • Apr 10, 2014
  • EURASIP Journal on Audio, Speech, and Music Processing
  • Aditya Arie Nugraha +2
  • Research Article
  • Citations27

Improvement of automatic speech recognition systems via nonlinear dynamical features evaluated from the recurrence plot of speech signals

  • Jul 12, 2016
  • Computers & Electrical Engineering
  • Shabnam Gholamdokht Firooz +2
  • Research Article
  • Citations15

Human-computer interaction for virtual-real fusion

  • Jan 01, 2023
  • Journal of Image and Graphics
  • Jianhua Tao +5
  • Research Article
  • Citations4

THE AUTOMATIC SPEAKER RECOGNITION SYSTEM OF CRITICAL USE CLASSIFIER OPTIMIZATION

  • Oct 04, 2018
  • Radio Electronics, Computer Science, Control
  • O V Bisikalo +2
  • Research Article
  • Citations2

A Lightweight Speaker Recognition System Using Timbre Properties

  • Jun 30, 2020
  • The Journal of Contents Computing
  • Abu Quwsar Ohi +5
  • Conference Article
  • Citations1

Selective MCE training strategy in Mandarin speech recognition

  • Nov 04, 2002
  • Jian-Min Zha +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.