Corpus ID: 40781814

Towards Goat Detection in Text-Dependent Speaker Verification

  title={Towards Goat Detection in Text-Dependent Speaker Verification},
  author={Orith Toledo-Ronen and H. Aronowitz and R. Hoory and Jason W. Pelecanos and D. Nahamoo},
We present a method that identifies speakers that are likely to have a high false-reject rate in a text-dependent speaker verification system (“goats”). The method normally uses only the enrollment data to perform this task. We begin with extracting an appropriate feature from each enrollment session. We then rank all the enrollment sessions in the system based on this feature. The lowest-ranking sessions are likely to have a high false-reject rate. We explore several features and show that the… Expand

Figures, Tables, and Topics from this paper

Detecting Goats in Speaker Verification Systems
We present a method for detecting goats in a textdependent speaker verification system using only the enrollment data. The goat detection process is based on extracting an appropriate feature fromExpand
RSR2015: Database for Text-Dependent Speaker Verification using Multiple Pass-Phrases
A protocol for the case of user-dependent pass-phrases in text-dependent speaker recognition is proposed and reported and speaker recognition experiments on RSR2015 database are reported. Expand
Text-dependent speaker verification: Classifiers, databases and RSR2015
The HiLAM system, based on a three layer acoustic architecture, and an i-vector/PLDA system, outperforms the state-of-the-art i- vector system in most of the scenarios and provides a reference evaluation scheme and a reference performance on RSR2015 database to the research community. Expand
New Developments in Voice Biometrics for User Authentication
This work investigates the use of state-of-the-art text-independent and text-dependent speaker verification technology for user authentication and shows how to adapt techniques such as joint factor analysis (JFA), Gaussian mixture models with nuisance attribute projection (GMMNAP), and hidden Markov models with NAP (HMM-NAP) to obtain improved results for new authentication scenarios and environments. Expand
Multi-modal biometrics for mobile authentication
A very high accuracy multi-modal authentication system based on fusion of several biometrics combined with a policy manager and a new biometric modality: chirography which is based on user writing on multi-touch screens using their finger is introduced. Expand


Model quality evaluation during enrolment for speaker verification
A new quality measure is proposed, which uses only data from clients, based on the number of training utterances that surpasses a certain threshold, which allows for the detection of non-representative utterances. Expand
A new procedure for classifying speakers in speaker verification systems
A new measure to classify speakers with respect to their behaviour in speaker recognition systems is proposed, and it is shown that measures based on a straightforward confusion matrices, that take only the 1-best classiication into account, cannot result in consistent classiications. Expand
Hunting for Wolves in Speaker Recognition
This work aims to predict which speaker pairs will be difficult for automatic speaker recognition systems to distinguish, by using features that characterize speakers, and thus provide a measure of speaker similarity, using data from NIST's 2008 Speaker Recognition Evaluation. Expand
Esca Workshop on Automatic Speaker Recognition, Identification and Verification the Pre-detection of Error-prone Class Members at the Enrollment Stage of Speaker Recognition Systems
| Two underlying factors that cause potential error-prone speakers in speaker veriication (SV) systems are examined, namely the inter-variance which indicates the individuality of the speaker, andExpand
SHEEP, GOATS, LAMBS and WOLVES: a statistical analysis of speaker performance in the NIST 1998 speaker recognition evaluation
This paper proposes statistical tests for the existence of sheep, goats, lambs and wolves and applies these tests to hunt for such animals using results from the 1998 NIST speaker recognition evaluation. Expand
On model quality and evaluation in speaker verification
While attempting to design measures for the quality of speaker models, a novel method for assigning weights to the contribution of models in accordance with their discriminative ability is developed. Expand
Simple and efficient speaker comparison using approximate KL divergence
A new approximate KL divergence distance extending earlier GMM parameter vector SVM kernels is used and a weighted nuisance projection method for channel compensation is applied, and a simple eigenvector method of training is presented. Expand
SVM Based Speaker Verification using a GMM Supervector Kernel and NAP Variability Compensation
A support vector machine kernel is constructed using the GMM supervector and similarities based on this kernel between the method of SVM nuisance attribute projection (NAP) and the recent results in latent factor analysis are shown. Expand
Support vector machines using GMM supervectors for speaker verification
This work examines the idea of using the GMM supervector in a support vector machine (SVM) classifier and proposes two new SVM kernels based on distance metrics between GMM models that produce excellent classification accuracy in a NIST speaker recognition evaluation task. Expand
Speaker verification in operational environments - monitoring for improved service operation
This paper presents a monitoring scheme for Speaker Verification dur­ ing the field test of a financial investment game and reveals a newly developed enrolment procedure, that can flag potentially weak speaker models, is an essential part of the monitoring procedure. Expand