A discriminative method for speaker verification using the difference information
نویسندگان
چکیده
In this paper, a discriminative method is proposed for speaker verification. An utterance can be mapped into a matrix by computing the difference to a codebook, and then expand the mapped matrix to a vector as the input of support vector machines for speaker verification. The Gaussian mixture modelbased method is also constructed by utilizing its nature. The mapped vector indicates the utterance's fitness to the codebook. Compared with the derivative operation in the famous fisher kernel the difference operation is used in our method. Experiments were run on the YOHO database in the textindependent case show that the new method is superior to the conventional GMM for speaker verification.
منابع مشابه
Using Exciting and Spectral Envelope Information and Matrix Quantization for Improvement of the Speaker Verification Systems
Speaker verification from talking a few words of sentences has many applications. Many methods as DTW, HMM, VQ and MQ can be used for speaker verification. We applied MQ for its precise, reliable and robust performance with computational simplicity. We also used pitch frequency and log gain contour for further improvement of the system performance.
متن کاملUsing Exciting and Spectral Envelope Information and Matrix Quantization for Improvement of the Speaker Verification Systems
Speaker verification from talking a few words of sentences has many applications. Many methods as DTW, HMM, VQ and MQ can be used for speaker verification. We applied MQ for its precise, reliable and robust performance with computational simplicity. We also used pitch frequency and log gain contour for further improvement of the system performance.
متن کاملComparison of discriminative training methods for speaker verification
The maximum likelihood estimation (MLE) and Bayesian maximum a-posteriori (MAP) adaptation methods for Gaussian mixture models (GMM) have proven to be effective and efficient for speaker verification, even though each speaker model is trained using only his own training utterances. Discriminative criteria aim at increasing discriminability by using out-of-class data. In this paper, we consider ...
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 کاملUnsupervised Discriminative Training of PLDA for Domain Adaptation in Speaker Verification
This paper presents, for the first time, unsupervised discriminative training of probabilistic linear discriminant analysis (unsupervised DT-PLDA). While discriminative training avoids the problem of generative training based on probabilistic model assumptions that often do not agree with actual data, it has been difficult to apply it to unsupervised scenarios because it can fit data with almos...
متن کامل