Vector quantization based Gaussian modeling for speaker verification

Gaussian mixture models (GMMs) have become an established means of modeling feature distributions in speaker recognition systems. It is useful for experimentation and practical implementation purposes to develop and test these models in an efficient manner particularly when computational resources are limited. A method of combining vector quantization (VQ) with single multi-dimensional Gaussians is proposed to rapidly generate a robust model approximation to the Gaussian mixture model. A fast method of testing these systems is also proposed and implemented. Results on the NIST 1996 Speaker Recognition Database suggest comparable and in some cases an improved verification performance to the traditional GMM based analysis scheme. In addition, previous research for the task of speaker identification indicated a similar system perfomance between the VQ Gaussian based technique and GMMs

An approach to statistical lip modelling for speaker identification via chromatic feature extraction

This paper presents a novel technique for the tracking of moving lips for the purpose of speaker identification. In our system, a model of the lip contour is formed directly from chromatic information in the lip region. Iterative refinement of contour point estimates is not required. Colour features are extracted from the lips via concatenated profiles taken around the lip contour. Reduction of order in lip features is obtained via principal component analysis (PCA) followed by linear discriminant analysis (LDA). Statistical speaker models are built from the lip features based on the Gaussian mixture model (GMM). Identification experiments performed on the M2VTS¹ database, show encouraging results

Robust speaker verification via fusion of speech and lip modalities

This paper investigates the use of lip information, in conjunction with speech information, for robust speaker verification in the presence of background noise. It has been previously shown in our own work, and in the work of others, that features extracted from a speaker's moving lips hold speaker dependencies which are complementary with speech features. We demonstrate that the fusion of lip and speech information allows for a highly robust speaker verification system which outperforms the performance of either sub-system. We present a new technique for determining the weighting to be applied to each modality so as to optimize the performance of the fused system. Given a correct weighting, lip information is shown to be highly effective for reducing the false acceptance and false rejection error rates in the presence of background noise

The use of temporal speech and lip information for multi-modal speaker identification via multi-stream HMMs

Investigates the use of temporal lip information, in conjunction with speech information, for robust, text-dependent speaker identification. We propose that significant speaker-dependent information can be obtained from moving lips, enabling speaker recognition systems to be highly robust in the presence of noise. The fusion structure for the audio and visual information is based around the use of multi-stream hidden Markov models (MSHMM), with audio and visual features forming two independent data streams. Recent work with multi-modal MSHMMs has been performed successfully for the task of speech recognition. The use of temporal lip information for speaker identification has been performed previously (T.J. Wark et al., 1998), however this has been restricted to output fusion via single-stream HMMs. We present an extension to this previous work, and show that a MSHMM is a valid structure for multi-modal speaker identification

The use of speech and lip modalities for robust speaker verification under adverse conditions

Investigates the use of lip information, in conjunction with speech information, for robust speaker verification in the presence of background noise. We have previously shown (Int. Conf. on Acoustics, Speech and Signal Proc., vol. 6, pp. 3693-3696, May 1998) that features extracted from a speaker's moving lips hold speaker dependencies which are complementary with speech features. We demonstrate that the fusion of lip and speech information allows for a highly robust speaker verification system which outperforms either subsystem individually. We present a new technique for determining the weighting to be applied to each modality so as to optimize the performance of the fused system. Given a correct weighting, lip information is shown to be highly effective for reducing the false acceptance and false rejection error rates in the presence of background noise

Manufacturing Yield Improvement by Clustering

Dealing with product yield and quality in manufacturing industries is getting more difficult due to the increasing volume and complexity of data and quicker time to market expectations. Data mining offers tools for quick discovery of relationships, patterns and knowledge in large databases. Growing self-organizing map (GSOM) is established as an efficient unsupervised datamining algorithm. In this study some modifications to the original GSOM are proposed for manufacturing yield improvement by clustering. These modifications include introduction of a clustering quality measure to evaluate the performance of the programme in separating good and faulty products and a filtering index to reduce noise from the dataset. Results show that the proposed method is able to effectively differentiate good and faulty products. It will help engineers construct the knowledge base to predict product quality automatically from collected data and provide insights for yield improvement.

i-vector based speaker recognition on short utterances

Robust speaker verification on short utterances remains a key consideration when deploying automatic speaker recognition, as many real world applications often have access to only limited duration speech data. This paper explores how the recent technologies focused around total variability modeling behave when training and testing utterance lengths are reduced. Results are presented which provide a comparison of Joint Factor Analysis (JFA) and i-vector based systems including various compensation techniques; Within-Class Covariance Normalization (WCCN), LDA, Scatter Difference Nuisance Attribute Projection (SDNAP) and Gaussian Probabilistic Linear Discriminant Analysis (GPLDA). Speaker verification performance for utterances with as little as 2 sec of data taken from the NIST Speaker Recognition Evaluations are presented to provide a clearer picture of the current performance characteristics of these techniques in short utterance conditions.

Identifying differences in safe roads and crash prone roads using clustering data mining

Road asset managers are overwhelmed with a high volume of raw data which they need to process and utilise in supporting their decision making. This paper presents a method that processes road-crash data of a whole road network and exposes hidden value inherent in the data by deploying the clustering data mining method. The goal of the method is to partition the road network into a set of groups (classes) based on common data and characterise the class crash types to produce a crash profiles for each cluster. By comparing similar road classes with differing crash types and rates, insight can be gained into these differences that are caused by the particular characteristics of their roads. These differences can be used as evidence in knowledge development and decision support.

Aggregate distance based clustering using Fibonacci Series -FIBCLUS

This paper proposes an innovative instance similarity based evaluation metric that reduces the search map for clustering to be performed. An aggregate global score is calculated for each instance using the novel idea of Fibonacci series. The use of Fibonacci numbers is able to separate the instances effectively and, in hence, the intra-cluster similarity is increased and the inter-cluster similarity is decreased during clustering. The proposed FIBCLUS algorithm is able to handle datasets with numerical, categorical and a mix of both types of attributes. Results obtained with FIBCLUS are compared with the results of existing algorithms such as k-means, x-means expected maximization and hierarchical algorithms that are widely used to cluster numeric, categorical and mix data types. Empirical analysis shows that FIBCLUS is able to produce better clustering solutions in terms of entropy, purity and F-score in comparison to the above described existing algorithms.

Clustering of web users using the tensor decomposed models

We propose to use the Tensor Space Modeling (TSM) to represent and analyze the user’s web log data that consists of multiple interests and spans across multiple dimensions. Further we propose to use the decomposition factors of the Tensors for clustering the users based on similarity of search behaviour. Preliminary results show that the proposed method outperforms the traditional Vector Space Model (VSM) based clustering.