• Journal of Convergence Society for SMB
• /
• v.6 no.4
• /
• pp.71-77
• /
• 2016

In this paper, we proposed a verification of speaker change utilizing the KL distance based on GMM-UBM to improve the performance of conventional BIC based Speaker Change Detection(SCD). We have verified Conventional BIC-based SCD using KL-distance based SCD which is robust against difference of information volume than BIC-based SCD. And we have applied GMM-UBM to compensate asymmetric information volume. Conventional BIC-based SCD was composed of two steps. Step 1, to detect the Speaker Change Candidate Point(SCCP). SCCP is positive local maximum point of dissimilarity d. Step 2, to determine the Speaker Change Point(SCP). If ${\Delta}BIC$ of SCCP is positive, it decides to SCP. We examined verification of SCP using GMM-UBM based KL distance D. If the value of D on each SCP is higher than threshold, we accepted that point to the final SCP. In the experimental condition MDR(Missed Detection Rate) is 0, FAR(False Alarm Rate) when the threshold value of 0.028 has been improved to 60.7%.

• Journal of Convergence for Information Technology
• /
• v.7 no.6
• /
• pp.159-166
• /
• 2017

In this paper, we have experimented to improve the verification performance of speaker change detection on broadcast news. It is to enhance the input noisy speech and to apply the KL distance $D_s$ using the SNR-based weighting function $w_m$. The basic experimental system is the verification system of speaker change using GMM-UBM based KL distance D(Experiment 0). Experiment 1 applies the input noisy speech enhancement using MMSE Log-STSA. Experiment 2 applies the new KL distance $D_s$ to the system of Experiment 1. Experiments were conducted under the condition of 0% MDR in order to prevent missing information of speaker change. The FAR of Experiment 0 was 71.5%. The FAR of Experiment 1 was 67.3%, which was 4.2% higher than that of Experiment 0. The FAR of experiment 2 was 60.7%, which was 10.8% higher than that of experiment 0.

• The Journal of the Acoustical Society of Korea
• /
• v.35 no.2
• /
• pp.83-91
• /
• 2016

Music-similarity computation is crucial in developing music information retrieval systems for browsing and classification. This paper overviews the recently-proposed centroid-model based music retrieval method and applies the distributional similarity measures to the model for retrieval-performance evaluation. Probabilistic distance measures (also called divergence) compute the distance between two probability distributions in a certain sense. In this paper, we consider the alpha divergence in computing distance between two centroid models for music retrieval. The alpha divergence includes the widely-used Kullback-Leibler divergence and Bhattacharyya distance depending on the values of alpha. Experiments were conducted on both genre and singer datasets. We compare the music-retrieval performance of the distributional similarity with that of the vector distances. The experimental results show that the alpha divergence improves the performance of the centroid-model based music retrieval.

• MALSORI
• /
• no.56
• /
• pp.225-235
• /
• 2005

Speech unit concatenation with a large database is presently the most popular method for speech synthesis. In this approach, the mismatches at the unit boundaries are unavoidable and become one of the reasons for quality degradation. This paper proposes an algorithm to reduce undesired discontinuities between the subsequent units. Optimal matching points are calculated in two steps. Firstly, the fullback-Leibler distance measurement is utilized for the spectral matching, then the unit sliding and the overlap windowing are used for the waveform matching. The proposed algorithm is implemented for the corpus-based unit concatenating Korean text-to-speech system that has an automatically labeled database. Experimental results show that our algorithm is fairly better than the raw concatenation or the overlap smoothing method.

• Journal of Computing Science and Engineering
• /
• v.7 no.2
• /
• pp.99-111
• /
• 2013

Approximating a divergence between two probability distributions from their samples is a fundamental challenge in statistics, information theory, and machine learning. A divergence approximator can be used for various purposes, such as two-sample homogeneity testing, change-point detection, and class-balance estimation. Furthermore, an approximator of a divergence between the joint distribution and the product of marginals can be used for independence testing, which has a wide range of applications, including feature selection and extraction, clustering, object matching, independent component analysis, and causal direction estimation. In this paper, we review recent advances in divergence approximation. Our emphasis is that directly approximating the divergence without estimating probability distributions is more sensible than a naive two-step approach of first estimating probability distributions and then approximating the divergence. Furthermore, despite the overwhelming popularity of the Kullback-Leibler divergence as a divergence measure, we argue that alternatives such as the Pearson divergence, the relative Pearson divergence, and the $L^2$-distance are more useful in practice because of their computationally efficient approximability, high numerical stability, and superior robustness against outliers.

• ETRI Journal
• /
• v.32 no.5
• /
• pp.795-800
• /
• 2010

Embedding a large vocabulary speech recognition system in mobile devices requires a reduced acoustic model obtained by eliminating redundant model parameters. In conventional optimization methods based on the minimum description length (MDL) criterion, a binary Gaussian tree is built at each state of a hidden Markov model by iteratively finding and merging similar mixture components. An optimal subset of the tree nodes is then selected to generate a downsized acoustic model. To obtain a better binary Gaussian tree by improving the process of finding the most similar Gaussian components, this paper proposes a new distance measure that exploits the difference in likelihood values for cases before and after two components are combined. The mixture weight of Gaussian components is also introduced in the component merging step. Experimental results show that the proposed method outperforms MDL-based optimization using either a Kullback-Leibler (KL) divergence or weighted KL divergence measure. The proposed method could also reduce the acoustic model size by 50% with less than a 1.5% increase in error rate compared to a baseline system.

• Communications for Statistical Applications and Methods
• /
• v.26 no.6
• /
• pp.623-633
• /
• 2019

This paper proposes two distance measures between two cumulative hazard functions that can be obtained by comparing their difference and ratio, respectively. Then we estimate the measures and present goodness of t test statistics. Since the proposed test statistics are expressed in terms of the cumulative hazard functions, we can easily give more weights on earlier (or later) departures in cumulative hazards if we like to place an emphasis on earlier (or later) departures. We also show that these test statistics present comparable performances with other well-known test statistics based on the empirical distribution function for an exponential null distribution. The proposed test statistic is an omnibus test which is applicable to other lots of distributions than an exponential distribution.

• Communications for Statistical Applications and Methods
• /
• v.21 no.5
• /
• pp.435-444
• /
• 2014

The most widely used optimal bandwidth is known to minimize the mean integrated squared error(MISE) of a kernel density estimator from a true density. In this article proposes, we propose a bandwidth which asymptotically minimizes the mean integrated density power divergence(MIDPD) between a true density and a corresponding kernel density estimator. An approximated form of the mean integrated density power divergence is derived and a bandwidth is obtained as a product of minimization based on the approximated form. The resulting bandwidth resembles the optimal bandwidth by Parzen (1962), but it reflects the nature of a model density more than the existing optimal bandwidths. We have one more choice of an optimal bandwidth with a firm theoretical background; in addition, an empirical study we show that the bandwidth from the mean integrated density power divergence can produce a density estimator fitting a sample better than the bandwidth from the mean integrated squared error.

• Journal of the Korea Society of Computer and Information
• /
• v.3 no.2
• /
• pp.131-138
• /
• 1998

In usual statistical data analysis, we describe statistical data by exact values. However, in modem complex and large-scale systems, it is difficult to treat the systems using only exact data. In this paper, we define these data as fuzzy data(ie. Linguistic variable applied to make the member-ship function.) and Propose a new method to get an analysis of fuzzy survey data based on the maximum entropy Principle. Also, we propose a new method of discrimination by measuring distance between a distribution of the stable state and estimated distribution of the present state using the Kullback - Leibler information. Furthermore, we investigate the validity of our method by computer simulations under realistic situations.

• Speech Sciences
• /
• v.10 no.3
• /
• pp.225-239
• /
• 2003

In this paper, we develop a keyword spotting system using vocabulary-independent speech recognition technique, and investigate several non-keyword modeling and post-processing methods to improve its performance. In order to model non-keyword speech segments, monophone clustering and Gaussian Mixture Model (GMM) are considered. We employ likelihood ratio scoring method for the post-processing schemes to verify the recognition results, and filler models, anti-subword models and N-best decoding results are considered as an alternative hypothesis for likelihood ratio scoring. We also examine different methods to construct anti-subword models. We evaluate the performance of our system on the automatic telephone exchange service task. The results show that GMM-based non-keyword modeling yields better performance than that using monophone clustering. According to the post-processing experiment, the method using anti-keyword model based on Kullback-Leibler distance and N-best decoding method show better performance than other methods, and we could reduce more than 50% of keyword recognition errors with keyword rejection rate of 5%.