A comparative evaluation of variance flooring techniques in HMM-based speaker verification
نویسندگان
چکیده
The problem of how to train variance parameters on scarce data is addressed in the context of text-dependent, HMM-based, automatic speaker verification. Three variations of variance flooring is explored as a means to prevent over-fitting. With the best performing one, the floor to a variance vector of a client model is proportional to the corresponding variance vector in a non-client multi-speaker model. It is also found that adapting the means and mixture weights from the non-client model while keeping variances constant works comparably to variance flooring and is much simpler. Comparisons are made on three large telephone quality corpora.
منابع مشابه
Advances on HMM-based text-dependent speaker verification
This paper presents recent development on text-dependent speaker verification technology in EU project PICASSO, which have improved the SV performance significantly. In the project we adopt HMM approach for pattern matching. In the paper we describes four different techniques, adaptive variance flooring, multiple use of enrolment sample, generalised competitive measurement for score normalisati...
متن کاملAn implementation and evaluation of an on-line speaker verification system for field trials
This paper presents a HMM-based speaker verification system which was implemented for a field trial. One of the challenges for moving HMM from speech recognition to speaker verification is to understand the HMM score variation and to define a proper measurement which is comparable across speech samples. In this paper we define two basic verification measurements, a qualifier-based measurement a...
متن کاملVariance flooring, scaling and tying for text-dependent speaker verification
The problem of how to estimate variance parameters in client models from scarce data is addressed in the context of textdependent, HMM-based, automatic speaker verification. Variance flooring and variance scaling are investigated as two alternative estimation techniques and are used with or without variance tying on the state level to reduce the number of parameters to estimate. The best result...
متن کاملOptimizing Variance Flooring in Hmm-based Speaker Verification
Adaptive variance flooring has previously been suggested as a method to help preventing over-fitting of variance parameters during training of HMMs. With this method, the average variance over some calibration data set is multiplied by a constant variance flooring factor to produce the variance floor. In this paper, experiments with adaptive variance flooring is presented, in which we try to op...
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 کامل