S. Ullah

University of Waterloo, Waterloo, Ontario, Canada

Are you S. Ullah?

Claim your profile

Publications (8)0 Total impact

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, a new method is proposed based on the side information and non-dominated sorting evolution strategy (NSES)-based K-means clustering algorithm. In a distance metric learning approach, data points are transformed to a new space where the Euclidean distances between similar and dissimilar points are at their minimum and maximum, respectively. However, the NSES-based K-means clustering yields globally optimized Gaussian components for an accent classification system. This hybrid clustering and classification approach enhances the performance of natural language call-routing systems. Accent classification performs the task of acoustic model switching based on the confidence measure for the callerpsilas query.
    Pattern Recognition, 2008. ICPR 2008. 19th International Conference on; 01/2009
  • S. Ullah, F. Karray
    [Show abstract] [Hide abstract]
    ABSTRACT: A speaker's accent is the most important factor affecting the performance of automatic speech recognition (ASR) systems because accents vary widely, even within the same country or community. This variation is due to the fact that when non- native speakers start to learn a second language, the substitution of native language phoneme pronunciation is a common process. Such substitution leads to fuzziness between the phoneme boundaries and phoneme classes. This fuzziness reduces out-of class variations and increases the similarities between the different sets of phonemes. In this paper, a new method is proposed based on the side information from dissimilar pairs of accent groups, to transfer data points to a new space where the Euclidian distances between similar and dissimilar points become minimum and maximum, respectively.
    Signal Processing and Information Technology, 2007 IEEE International Symposium on; 01/2008
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper proposes a multi-objective evolution strategy (ES) hybridized with a k-means algorithm to address a data clustering problem whose objective is minimizing both clustering error and cluster number. Contrary to the conventional data clustering problem with a predetermined number of clusters, the bi-objective problem considered in this study has a set of clustering solutions whose cluster numbers are different from one another. This enables to secure the best clustering result that fits specific needs without restricting the cluster number. To find the solution set, the hybrid ES evolves a population of solution candidates each of which represents a variable number of cluster centroids. While evolving the population, special ES operators dedicated to the bi-objective clustering problem are used. Whenever the hybrid ES creates a new set of cluster centroids, it is fine-tuned by the k-means algorithm. The experiment results show that the hybrid ES outperforms the conventional ES and KMA.
    Control, Automation and Systems, 2008. ICCAS 2008. International Conference on; 01/2008
  • Sameeh Ullah, Fakhri Karray
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper describes a speaker-independent accent-based natural language call-routing system. Based on a speaker's accent group, this system directs customer calls to the automatic speech recognition system that is most suitable to recognize the input query. The speech recognition system understands the caller's query and converts it into routing keywords. Accent identification is the most important factor for improving the performance of natural language call-routing systems because accents vary widely, even within the same country or community. This variation occurs when non-native speakers start to learn a second language; the substitution of native language phoneme pronunciation is a common occurrence. In this paper, a new method is proposed based on class inequivalent side information and an evolutionary-based K-means clustering algorithm. In a distance metric learning approach, data points are transferred to a new space where the Euclidean distances between similar and dissimilar points are at their minimum and maximum, respectively. However, the evolutionary-based K-means clustering approach yields globally optimized Gaussian components for an accent classification system.
    01/2008;
  • S. Ullah, F. Karray
    [Show abstract] [Hide abstract]
    ABSTRACT: A speaker's accent is the most important factor affecting the performance of automatic speech recognition (ASR) systems. This is due to the fact that accents vary widely, even within the same country or community. The reason may be attributed to the fuzziness between the boundaries of phoneme classes, a result of differences in a speaker's vocal tract and accent. In this paper, a new method of accent classification is proposed that is based on fuzzy Gaussian mixture models (FGMMs). The proposed method first uses a fuzzy clustering to fuzzily partition the data. In this way, fuzzy memberships to the cluster centres are determined by minimizing the distance between the cluster centres and feature vectors. Afterwards, a GMM classifier is trained by using the fuzzy Gaussian parameters to classify the speaker's accent. The experimental results show that the proposed method outperforms the Gaussian Mixture models, Vector Quantization modeling method, Hidden Markov Model, and Radial Basis Neural Networks.
    Information and Emerging Technologies, 2007. ICIET 2007. International Conference on; 08/2007
  • S. Ullah, F. Karray
    [Show abstract] [Hide abstract]
    ABSTRACT: Call routing based on natural language identification is an important issue for call centres operating in a multilingual scenario. This is due to the fact that it is not possible for a human agent to become fluent in all languages. In this paper, we propose a call routing system based on prosodic and phonetic features to improve the performance of automatic call routing systems. The main focus of this paper is to propose a feature set that efficiently improves the performance of the system without incorporating several sets of features. Our proposed approach for combining the prosodic and phonetic features achieves an accuracy rate of 98.36% for a binary language identification call routing system.
    Information and Emerging Technologies, 2007. ICIET 2007. International Conference on; 08/2007
  • [Show abstract] [Hide abstract]
    ABSTRACT: Call routing based on Natural Language Understanding remains a complex and challenging research area in machine intelligence and language understanding. This is despite the apparent limited success of a few commercial natural language call routing systems. This challenge is due to the limitations imposed by the speech recognition engine, the language model, and the natural language parser. In this paper, we propose a system to enhance the performance of automated call routing applications based on knowledge-based networks. The main focus of this paper is on the enhancement of the performance at the Natural Language Understanding level. We investigate soft computing techniques such as Learning Vector Quantization and the Genetic Algorithm. We find that the Genetic Algorithm outperforms Learning Vector Quantization. We achieve an accuracy rate of 84.12% using the Genetic Algorithm vs. 73% for Learning Vector Quantization.
    Signal Processing and Its Applications, 2007. ISSPA 2007. 9th International Symposium on; 03/2007
  • S. Ullah, F. Karray
    [Show abstract] [Hide abstract]
    ABSTRACT: A speaker's accent is the most important factor affecting the performance of automatic speech recognition (ASR) systems because accents vary widely, even within the same country or community. The reason may be attributed to the fuzziness between the boundaries of phoneme classes, a result of differences in a speaker's vocal tract and accent. In this paper, a new method is proposed that is based on the fuzzy canonical correlation-based Gaussian classifier. In the proposed method the membership values of the clusters are based not only on minimizing the distance between the cluster centroids, but also maximizing the out-of-class variations.
    01/2007;

Publication Stats

5 Citations

Top Journals

Institutions

  • 2007–2009
    • University of Waterloo
      • Department of Electrical & Computer Engineering
      Waterloo, Ontario, Canada