Identification and Recognition of Speaker Voice Using a Neural Network-Based Algorithm: Deep Learning

Identification and Recognition of Speaker Voice Using a Neural Network-Based Algorithm: Deep Learning

Neeraja Koppula, K. Sarada, Ibrahim Patel, R. Aamani, K. Saikumar
DOI: 10.4018/978-1-7998-6870-5.ch019
OnDemand:
(Individual Chapters)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

This chapter explains the speech signal in moving objects depending on the recognition field by retrieving the name of individual voice speech and speaker personality. The adequacy of precisely distinguishing a speaker is centred exclusively on vocal features, as voice contact with machines is getting more pervasive in errands like phone, banking exchanges, and the change of information from discourse data sets. This audit shows the location of text-subordinate speakers, which distinguishes a solitary speaker from a known populace. The highlights are eliminated; the discourse signal is enrolled for six speakers. Extraction of the capacity is accomplished utilizing LPC coefficients, AMDF computation, and DFT. By adding certain highlights as information, the neural organization is prepared. For additional correlation, the attributes are put away in models. The qualities that should be characterized for the speakers were acquired and dissected utilizing back propagation algorithm to a format picture.
Chapter Preview
Top

1 Introduction

Expanding request, brought about by the advancement of urban areas and developing populace, draws more consideration on power transmission limits. Force Electronics assists with tackling issue by presenting adaptable air conditioning transmission framework (FACTS) gadgets. The general favorable circumstances of FACTS gadgets are notable. In any case, FACTS gadgets can prompt extreme issues in deciding the specific area of the shortcoming when they are remembered for the deficiency zone. Separation transfer's capacity depends on estimated impedance by that they compute flaw area. Introduced FACTS gadgets on TL change the impedance estimated by separation hand-off that thusly prompts wrong flaw zone location. The measure of deviation in impedance relies upon the boundary's settings of FACTS gadget.

The wavelet change is utilized to isolate high-recurrence parts of 3-line flows, and help vector machine (utilizing techniques for multi- class utilization) is utilized for order of issue area into three assurance locales of separation transfer.

Lately, specialists have gotten progressively keen on comprehending this issue. For demonstrating arrangement capacitor in the shortcoming time frame, the greater part of articles utilized model introduced by Goldsworthy Goldsworthy (1987). In Sadeh, Javad (2020) an answer dependent on time conditions of system with thought of a broad method of TL is introduced. In this arrangement, zone and area of deficiency, autonomous of issue obstruction, are evaluated utilizing 2 subroutines. A few articles built up their strategies utilizing estimations from rule segments of primary recurrence. Saha et al(1999) built up a strategy dependent on segments of principle recurrence on 1 side of TL utilizing Goldsworthy method.

In Song, Y. H(1996), a calculation dependent on forward and in reverse waves is introduced to secure arrangement repaid TLs; albeit just one stage to ground deficiency are examined in this manuscript. Information is introduced in Thomas, D (1992) and Xuan,Q.Y (1996) that recommend strategies dependent on neural systems. It is proposed to utilize RBFN as a spiral neural system to order flaw kind and issue area Girgis, Adly A (1998) and Kalman,(1960). The work Megahed(2006),trains 2 unique kinds of neural systems to distinguish deficiency kind and area. Likewise, Dash, P. K (2206) utilizes EDBD calculation to figure deficiency area Fattaheian, Dehkordi(2014). Neural systems have impediments, incorporating their requirement for countless neurons to demonstrate organize structure (2008). Along these lines, various information from the electrical system is required, and furthermore much an ideal opportunity to prepare the system. By and by, these articles utilized a predetermined number of information to test their techniques. The work Samantaray, S. R (2013) suggested a strategy dependent on Kalman channel] that gauges the condition of a powerful framework from a progression of estimations comprisingblunders.

Complete Chapter List

Search this Book:
Reset