Text-independent speaker verification using rank threshold in large number of speaker models
نویسندگان
چکیده
In this paper, we propose a novel speaker verification method which determines whether a claimer is accepted or rejected by the rank of the claimer in a large number of speaker models instead of score normalization, such as T-norm and Z-norm. The method has advantages over the standard T-norm in speaker verification accuracy. However, it needs much computation time as well as T-norm that needs calculating likelihoods for many cohort models. Hence, we also discuss the speed-up using the method that selects cohort subset for each target speaker in the training stage. This data driven approach can significantly reduce computation resulting in faster speaker verification decision. We conducted text-independent speaker verification experiments using large-scale Japanese speaker recognition evaluation corpus constructed by National Research Institute of Police Science. As a result, the proposed method achieved an equal error rate of 2.2 %, while T-norm obtained 2.7 %.
منابع مشابه
Determination of threshold for speaker verification using speaker adaptation gain in likelihood during training
This paper describes methods to determine thresholds for speaker verification. Setting an appropriate threshold a priori is difficult because likelihood verification covers a wide range and the appropriate threshold for each speaker is different. We propose new methods to determine the speaker verification threshold depending on the "adaptation degree" for each speaker. We use the gain in likel...
متن کاملComparison of background normalization methods for text-independent speaker verification
This paper compares two approaches to background model representation for a text-independent speaker verification task using Gaussian mixture models. We compare speaker-dependent background speaker sets to the use of a universal, speaker-independent background model (UBM). For the UBM, we describe how Bayesian adaptation can be used to derive claimant speaker models, providing a structure leadi...
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 کاملNew background modeling for speaker verification
A new background speaker modelling method is presented in this paper for text-independent speaker verification using Gaussian mixture models. This method does not require speech databases of other speakers to build background speaker models. A background model can be built directly from the same claimed speaker's database and has a smaller number of Gaussian mixtures compared to the claimed spe...
متن کامل