Pertinent Prosodic Features for Speaker Identification by Voice

Pertinent Prosodic Features for Speaker Identification by Voice

Halim Sayoud (USTHB University, Algeria) and Siham Ouamour (USTHB University, Algeria)
DOI: 10.4018/978-1-4666-0119-2.ch015
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Most existing systems of speaker recognition use “state of the art” acoustic features. However, many times one can only recognize a speaker by his or her prosodic features, especially by the accent. For this reason, the authors investigate some pertinent prosodic features that can be associated with other classic acoustic features, in order to improve the recognition accuracy. The authors have developed a new prosodic model using a modified LVQ (Learning Vector Quantization) algorithm, which is called MLVQ (Modified LVQ). This model is composed of three reduced prosodic features: the mean of the pitch, original duration, and low-frequency energy. Since these features are heterogeneous, a new optimized metric has been proposed that is called Optimized Distance for Heterogeneous Features (ODHEF). Tests of speaker identification are done on Arabic corpus because the NIST evaluations showed that speaker verification scores depend on the spoken language and that some of the worst scores were got for the Arabic language. Experimental results show good performances of the new prosodic approach.
Chapter Preview
Top

Database

We used an Arabic speech database which contains a lot of pertinent phonemes (a pertinent phoneme is a phoneme which characterize well the speakers) (Sayoud & Ouamour, 2004). These phonemes are given in table 1.

Table 1.
Some pertinent Arabic phonemes used in our sentences
Phoneme designationLatin similarityArabic Transcription
“Kaf”  K
“Rra”  Spanish R
“Aïn”  Nothing
“Dtta”  Nothing

Complete Chapter List

Search this Book:
Reset