Short Utterance Speaker Recognition A research Agenda

Fatima, Nakhat; Zheng, Thomas Fang

doi:10.1109/icsai.2012.6223381

Cited by 20 publications

(6 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A challenge for these domains outside of psychology is in developing short utterance speaker recognition systems in order to facilitate speaker recognition when only small amounts of data are available (Fatima and Fang Zheng, 2012). Here, we investigated the cognitive mechanisms that enable efficient speaker recognition from very short voice samples.…”

Section: Discussionmentioning

confidence: 99%

Familiarity and Voice Representation: From Acoustic-Based Representation to Voice Averages

2017

View full text Add to dashboard Cite

The ability to recognize an individual from their voice is a widespread ability with a long evolutionary history. Yet, the perceptual representation of familiar voices is ill-defined. In two experiments, we explored the neuropsychological processes involved in the perception of voice identity. We specifically explored the hypothesis that familiar voices (trained-to-familiar (Experiment 1), and famous voices (Experiment 2)) are represented as a whole complex pattern, well approximated by the average of multiple utterances produced by a single speaker. In experiment 1, participants learned three voices over several sessions, and performed a three-alternative forced-choice identification task on original voice samples and several “speaker averages,” created by morphing across varying numbers of different vowels (e.g., [a] and [i]) produced by the same speaker. In experiment 2, the same participants performed the same task on voice samples produced by familiar speakers. The two experiments showed that for famous voices, but not for trained-to-familiar voices, identification performance increased and response times decreased as a function of the number of utterances in the averages. This study sheds light on the perceptual representation of familiar voices, and demonstrates the power of average in recognizing familiar voices. The speaker average captures the unique characteristics of a speaker, and thus retains the information essential for recognition; it acts as a prototype of the speaker.

show abstract

Section: Discussionmentioning

confidence: 99%

Familiarity and Voice Representation: From Acoustic-Based Representation to Voice Averages

2017

View full text Add to dashboard Cite

show abstract

“…They were improved with new techniques like in [10], eigenvoices, principal component analysis or LDA (linear discriminant analysis) [15]. Recently, support vector machine (SVM)‐based technologies are used [3, 16]. Generally, these systems lead to comparable or superior performances than the other systems of classification in our domain [17, 18].…”

Section: Previous Workmentioning

confidence: 99%

“…In practise, SR requires a large amount of speech data, making use of huge files and complicated processing [2]. This has hampered the SR technology to be used widely [3] especially for a realistic application which is limited by constraints related to the memory and computational resource limitation or even the utterance duration fixed by the system. Embedded systems which involve relatively short duration speech material might perfectly present these conditions [4].…”

Section: Introductionmentioning

confidence: 99%

New approach for short utterance speaker identification

Chakroun

Frikha

Zouari

2018

IET signal process.

View full text Add to dashboard Cite

Recent advances in the speaker recognition (SR) field showed remarkably accurate and outperforming algorithms. However, their performances drastically degrade when the sparse amount of data is available. Nowadays, recognising a speaker identity when only a small amount of speech data is involved for testing and training remains a key consideration since many real world applications often have access to only speech data having a limited duration. In this study, the authors present a new improved approach, based on new information detected from the speech signal, to improve the task of automatic speaker identification. In doing so, they highlight how the detection of the speaker dialect can be explored to address the research problem related to short utterance SR. Results obtained with the new regional system are presented which provide a comparison between this system and the state-of-the-art systems for speaker identification task.

show abstract

“…On the other hand, the purpose of achieving verification process is to confirm the person to be true or imposter [6].…”

Section: Speaker Recognitionmentioning

confidence: 99%

Distance Method with PCA based Speaker Identification

Khaled¹

2016

IJCA

View full text Add to dashboard Cite

This paper is dedicated to present statistic method with dimensionality reduction for speaker identification. LPC is involved to formed set of descriptors for the speaker data. The formed descriptors vectors are evaluated to realize dimensionality reduction by PCA. This set of selected features is recognized by distance method. The experimental result revealed that the proposed work is achieved in better performance than without reduction.

show abstract

Short Utterance Speaker Recognition A research Agenda

Cited by 20 publications

References 14 publications

Familiarity and Voice Representation: From Acoustic-Based Representation to Voice Averages

Familiarity and Voice Representation: From Acoustic-Based Representation to Voice Averages

New approach for short utterance speaker identification

Distance Method with PCA based Speaker Identification

Contact Info

Product

Resources

About