A study on model-based equal error rate estimation for automatic speaker verification
نویسندگان
چکیده
Usually, we need a large number of testing samples to evaluate the performance of automatic speaker verification (ASV) systems. The equal error rate (EER) is a common measure for this purpose. It is derived according to the threshold determined by finding the verification score when the false rejection rate (FRR) equals to the false acceptance rate (FAR). In this paper, a method of model-based EER estimation for the ASV system is proposed. The goal is to estimate the EER directly from the speaker model parameters without running the speaker verification experiments using a large number of testing samples. The verification scores are computed using the model parameters, and then both FRR and FAR are derived. With a small number of testing samples, we can adjust the score distribution to estimate the EER of the ASV system. The experimental result shows that the proposed method is effective and very promising for feedback loop design of an ASV system.
منابع مشابه
Using Vector Quantization for Universal Background Model in Automatic Speaker Verification
We aim to describe different approaches for vector quantization in Automatic Speaker Verification. We designed our novel architecture based on multiples codebook representing the speakers and the impostor model called universal background model and compared it to another vector quantization approach used for reducing training data. We compared our scheme with the baseline system, Gaussian Mixtu...
متن کاملAnti-spoofing Methods for Automatic Speaker Verification System
Growing interest in automatic speaker verification (ASV) systems has lead to significant quality improvement of spoofing attacks on them. Many research works confirm that despite the low equal error rate (EER) ASV systems are still vulnerable to spoofing attacks. In this work we overview different acoustic feature spaces and classifiers to determine reliable and robust countermeasures against s...
متن کاملDiscriminative adaptation for speaker verification
Speaker verification is a binary classification task to determine whether a claimed speaker uttered a phrase. Current approaches to speaker verification tasks typically involve adapting a general speaker Universal Background Model (UBM), normally a Gaussian Mixture Model (GMM), to model a particular speaker. Verification is then performed by comparing the likelihoods from the speaker model to t...
متن کاملOptimizing feature complementarity by evolution strategy: Application to automatic speaker verification
Conventional automatic speaker verification systems are based on cepstral features like Mel-scale Frequency Cepstrum Coefficient (MFCC), or Linear Predictive Cepstrum Coefficient (LPCC). Recent published works showed that the use of complementary features can significantly improve the system performances. In this paper, we propose to use an evolution strategy to optimize the complementarity of ...
متن کاملClose Speakers Model and Comparative Study in Automatic Speaker Verification
The performance of speaker verification system degrades when the test segments are utterances of short duration, therefore, we investigate the use of model representing our target speaker with his close speaker and his own speech data. We propose to create a new Speaker Model who groups close speakers (CS) achieved with two clustering algorithms in Automatic Speaker Verification A.S.V. Intra an...
متن کامل