Verification effectiveness in open-set speaker identification

Ariyaeeinia, Aladdin M.; Fortuna, J.; Sivakumaran, P.; Malegaonkar, A.

doi:10.1049/ip-vis:20050273

Cited by 21 publications

(38 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The cohort size of the competing models is set to 1 and 3 in the cases of clean and degraded data, respectively. This is in agreement with the findings in some earlier studies [9,10]. The procedures for speech feature extraction and speaker classification are as detailed in Refs.…”

Section: Experimental Investigations and Resultssupporting

confidence: 92%

“…The procedures for speech feature extraction and speaker classification are as detailed in Refs. [10,12]. The face recognition scores are based on the approaches detailed in Refs.…”

Section: Experimental Investigations and Resultsmentioning

confidence: 99%

“…However, since the observed test material is in fact the test utterance, what is required to be computed is the probability of the target model, given the test utterance. These two properties are related through the Bayes' theorem as [9,10] …”

Section: Motivation and Proposed Approachmentioning

confidence: 99%

“…p(x), on the other hand, will need to be approximated. To date, there have been different approximation approaches introduced for this purpose, leading to different score normalisation methods [9][10][11]. A slightly different approach to score normalisation in speaker recognition is that based on the standardisation of score distributions, with the aim to facilitate the use of a single threshold for all registered speakers [8].…”

Section: Motivation and Proposed Approachmentioning

confidence: 99%

“…The different methods in this two category of score normalisation (i.e. Bayesian and standardisation) have already been subjected to thorough comparative evaluations in the context of speaker recognition [10,12]. The normalisation methods considered for this purpose are cohort normalisation (CN), unconstrained cohort normalisation (UCN), universal background model (UBM) normalisation, T-norm and Z-norm.…”

Section: Motivation and Proposed Approachmentioning

confidence: 99%

See 4 more Smart Citations

Enhancement of multimodal biometric segregation using unconstrained cohort normalisation

Alsaade

Ariyaeeinia

Malegaonkar

et al. 2008

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

This paper presents an investigation into the effects, on the accuracy of multimodal biometrics, of introducing unconstrained cohort normalisation (UCN) into the score-level fusion process. Whilst score normalisation has been widely used in voice biometrics, its effectiveness in other biometrics has not been previously investigated. This study aims to explore the potential usefulness of the said score normalisation technique in face biometrics and to investigate its effectiveness for enhancing the accuracy of multimodal biometrics. The experimental investigations involve the two recognition modes of verification and open-set identification, in clean mixed-quality and degraded data conditions. Based on the experimental results, it is demonstrated that the capabilities provided by UCN can significantly improve the accuracy of fused biometrics. The paper presents the motivation for, and the potential advantages of, the proposed approach and details the experimental study.

show abstract

Section: Experimental Investigations and Resultssupporting

confidence: 92%

“…The procedures for speech feature extraction and speaker classification are as detailed in Refs. [10,12]. The face recognition scores are based on the approaches detailed in Refs.…”

Section: Experimental Investigations and Resultsmentioning

confidence: 99%

Section: Motivation and Proposed Approachmentioning

confidence: 99%

Section: Motivation and Proposed Approachmentioning

confidence: 99%

Section: Motivation and Proposed Approachmentioning

confidence: 99%

See 3 more Smart Citations

Enhancement of multimodal biometric segregation using unconstrained cohort normalisation

Alsaade

Ariyaeeinia

Malegaonkar

et al. 2008

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

show abstract

Privacy‐preserving hands‐free voice authentication leveraging edge technology

Alattar¹,

Abbes²,

Zerai³

2022

Security and Privacy

View full text Add to dashboard Cite

Although speech recognition technology has improved significantly over the past few years, the absence of reliable voice authentication methods has negatively affected the Internet of Things (IoT). Voice-activated devices, by design, rely on personal voice biometrics for access-level security, and insecure voice authentication techniques limit the development of voice interfaces that can deal with multiple users with different privileges. Furthermore, public concerns surrounding biometric systems have been reported. Outsourcing biometric data are liable to various privacy breaches, increasing vulnerability to cyberattacks from both criminal organizations and government agencies. In this article, we address the security and privacy challenges of voice authentication for IoT by presenting an edge-based, hands-free, single-factor authentication scheme that consists of three security-based features: random phrase verification, anti-spoof verification, and text-independent speaker recognition. The use of edge technology allows the integration of this authentication scheme with resource contained IoT devices such as voice assistants. Additionally, privacy is ensured by a novel security protocol that uses chaffing and encryption techniques to protect the users' biometric data stored locally on the edge device. Finally, the scheme does not compromise the user experience as it requires only two short voice inputs.

show abstract

Performance Evaluation in Open-Set Speaker Identification

Malegaonkar¹,

Ariyaeeinia

2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. The concern in this study is the approach to evaluating the performance of the open-set speaker identification process. In essence, such a process involves first identifying the speaker model in the database that best matches the given test utterance, and then determining if the test utterance has actually been produced by the speaker associated with the best-matched model. Whilst, conventionally, the performance of each of these two sub-processes is evaluated independently, it is argued that the use of a measure of performance for the complete process can provide a more useful basis for comparing the effectiveness of different systems. Based on this argument, an approach to assessing the performance of open-set speaker identification is considered in this paper, which is in principle similar to the method used for computing the diarisation error rate. The paper details the above approach for assessing the performance of open-set speaker identification and presents an analysis of its characteristics.

show abstract

Verification effectiveness in open-set speaker identification

Cited by 21 publications

References 12 publications

Enhancement of multimodal biometric segregation using unconstrained cohort normalisation

Enhancement of multimodal biometric segregation using unconstrained cohort normalisation

Privacy‐preserving hands‐free voice authentication leveraging edge technology

Performance Evaluation in Open-Set Speaker Identification

Contact Info

Product

Resources

About