scispace - formally typeset
Search or ask a question

Showing papers on "Speaker recognition published in 1974"


Journal ArticleDOI
J.-P. Haton1
TL;DR: An application of a real-time isolated-word recognition system, to the recognition of sentences of a language used in numerical command of machine tools, which will then be adapted to the Recognition of connected speech, as well as in the design of new languages for man-machine voice communication.
Abstract: In this paper we describe an application of a real-time isolated-word recognition system, to the recognition of sentences of a language used in numerical command of machine tools. The acoustic level operates with dynamic matching procedure and knowledge about syntactics and semantics of the language is used to predict the incoming words. With such a syntax-directed system, real-time recognition of sentences pronounced word-by-word is very accurately achieved, even for several speakers. This system will then be adapted to the recognition of connected speech, as well as in the design of new languages for man-machine voice communication.

16 citations


Journal ArticleDOI
TL;DR: The hypothesis of no difference between the recognition of sung and spoken voices was tested by presenting 24 Ss with two dichotic voice recognition tests which were similar in all respects except that stimuli were spoken on one and sung on the other.
Abstract: The hypothesis of no difference between the recognition of sung and spoken voices was tested by presenting 24 Ss with two dichotic voice recognition tests which were similar in all respects except that stimuli were spoken on one and sung on the other. A finding of significantly more accurate speaker recognition indicated differences in the processing of singing and speaking voices. Results of the singer recognition test replicated a previous finding of no significant difference between ears. A similar lack of lateralization for speaker recognition could not be unambiguously interpreted because the absence of ear asymmetry was significantly related to very high levels of overall accuracy.

9 citations





Journal ArticleDOI
TL;DR: In this article, a speaker identification technique that extracts speaker characteristics from single sound units can be applicable to both text-dependent and text-independent speech is presented. But, there is no control over the number of sound units in the test speech sample, so it would be beneficial if the probability of success could be predicted not only from the population size but also from the number and type of sound unit available.
Abstract: Speaker identification techniques which extract speaker characteristics from single sound units can be applicable to both text‐dependent and text‐independent speech. Experiments have shown that identification accuracy improves if decisions are based upon combined information from several different sound units. With such techniques, it would be helpful if potential sound units could be rank ordered according to how much each contributes toward successful speaker identification. In text‐independent speaker identification there is no control over the number of sound units in the test speech sample, so it would be beneficial if the probability of success could be predicted not only from the population size but also from the number and type of sound units available. Speech samples from ten male speakers are being used to evaluate ten non‐diphthongized vowels and three nasals. Experimental results are presented on rank ordering these sounds and the determination of conditional probabilities based upon the selection of sounds. [Sponsored by Air Force Systems Command, Rome Air Development Center, Griffiss Air Force Base, New York 13441.]

1 citations


Journal ArticleDOI
TL;DR: In this article, the pitch and duration of the vowels for different speakers were plotted in a two-dimensional space for each context, indicating different contours for the speakers, and the separation between the contours is even better when the data for different contexts are pooled.
Abstract: This paper reports some experimental studies related to the possibility of using sonagrams (or “Voiceprints”) for recognizing the speaker. In the present experiment ten (Hindi) vowels, embedded in four different contexts viz. isolation, h-d, hΛ y-y (concavity), and hΛ w-w (convexity), as spoken by three male speakers were analyzed with the ‘sonograph’. Both the fundamental frequency and the duration of the vowels were estimated. The data indicates a significant variation in these two parameters as a function of speaker as well as context. The pitch and duration of the vowels for different speakers were plotted in a two-dimensional space for each context. The plots (or vowel contours) in all the contexts indicate different contours for the speakers. The separation between the contours is even better when the data for different contexts are pooled. These observations suggest that pitch-duration contours of vowels could serve as satisfactory invarient cues for individual speakers.