The Experts below are selected from a list of 273 Experts worldwide ranked by ideXlab platform

Wushour Silamu - One of the best experts on this subject based on the ideXlab platform.

  • CSIE (7) - HMM-Based Uyghur Continuous Speech Recognition System
    2009 WRI World Congress on Computer Science and Information Engineering, 2009
    Co-Authors: Wushour Silamu, Nasirjan Tursun
    Abstract:

    In this work presents a Continuous Speech recognition system for Uyghur language based on HMM, which called UASRS. Uyghur language is an agglutinative language and one of the least studied languages on Speech recognition area. So, our first work was building a Uyghur Continuous Speech database. In acoustic level, we was using the common used HMM (Hidden Marcov Model) for modeling the Uyghur Speech data; in language level, modeling the Uyghur text data based on N-Gram language model. At last we were using the recognizer of HTK3.3 (HMM ToolKit) and the MS Visual C++8.0 developing the Uyghur Continuous Speech Recognition System. In this paper also presents the recognition experiments of Uyghur Continuous Speech by using the UASRS. The recognition rate was 68.98% (sentences), and 94.65% (words) for the test set. The recognition rate was 51.49% (sentences), and 85.82% (words) for the real-time Speech recognition.

  • Uyghur Continuous Speech recognition system based on HMM
    Journal of Computer Applications, 2009
    Co-Authors: Wushour Silamu
    Abstract:

    Uyghur language is an agglutinative language.It is possible to produce a very large number of words from the same root with suffixes,so that the Speech recognition of Uyghur language is very difficult.Combined with the characteristics of Uyghur language,this paper built a Uyghur Continuous Speech database,and designed the Hidden Markov Model(HMM) based Uyghur Continuous Speech recognition system by using the HTK(HMMToolKit).On the acoustic level,this paper selected triphone as the basic recognition unit,and used many methods such as decision tree,tied-state triphones,fixing the silence models,increasing Gaussian mixture distribution to improve the precision of the models.On the language level,this study used the statistics-based bigram language model.Finally,this paper presented some recognition experiments by using that system.

Y. Nara - One of the best experts on this subject based on the ideXlab platform.

  • ICASSP - Interactive extraction of phonemic variation rules in Continuous Speech
    ICASSP '86. IEEE International Conference on Acoustics Speech and Signal Processing, 1
    Co-Authors: S. Kimura, Y. Nara
    Abstract:

    An approach to Continuous Speech recognition is introduced, and an extraction method for phonemic variation rules and the extracted rules are reported. To realize a Continuous Speech recognizer, we must solve the problem of phonemic variations in Continuous Speech. We use a top-down method. Our current effort is focused on interactive analysis of phonemic variations in Continuous Speech and extraction of the phonemic variation rules, to construct a data base of these rules. We analyzed phonemic variations in 1000 Japanese-language phrases spoken by a male speaker, and confirmed that all the phonemic variations in the 1000 phrases can be represented by about 160 rules. In addition, we obtained the occurrence probability of each phonemic variation from the frequency of use of each rule.

  • ICASSP - Extraction of phonemic variation rules in Continuous Speech spoken by multiple speakers
    ICASSP '87. IEEE International Conference on Acoustics Speech and Signal Processing, 1
    Co-Authors: S. Kimura, Y. Nara
    Abstract:

    This paper describes an interactive extraction of phonemic variation rules in Continuous Speech spoken by multiple speakers. To realize a Continuous Speech recognizer, we must first develop a highly accurate phoneme recognizer. The major problem related to phoneme recognizers is the phonemic variations in Continuous Speech. Our work focuses on the interactive analysis of phonemic variations in Continuous Speech and the extraction of the phonemic variation rules for many speakers. We extracted 317 rules related to 21 kinds of phonemic variation phenomena from 10,000 Japanese-language phrases spoken by 10 male speakers. With these rules, 97.6% of 36,000 Japanese-language phrases spoken by 36 test speakers (30 males and 6 females) were correctly segmented by our top-down phoneme segmentation system. Furthermore, a subset of the rules for each speaker was automatically obtained. On average, each subset contains 53.2% of the rules.

S. Kimura - One of the best experts on this subject based on the ideXlab platform.

  • ICASSP - Interactive extraction of phonemic variation rules in Continuous Speech
    ICASSP '86. IEEE International Conference on Acoustics Speech and Signal Processing, 1
    Co-Authors: S. Kimura, Y. Nara
    Abstract:

    An approach to Continuous Speech recognition is introduced, and an extraction method for phonemic variation rules and the extracted rules are reported. To realize a Continuous Speech recognizer, we must solve the problem of phonemic variations in Continuous Speech. We use a top-down method. Our current effort is focused on interactive analysis of phonemic variations in Continuous Speech and extraction of the phonemic variation rules, to construct a data base of these rules. We analyzed phonemic variations in 1000 Japanese-language phrases spoken by a male speaker, and confirmed that all the phonemic variations in the 1000 phrases can be represented by about 160 rules. In addition, we obtained the occurrence probability of each phonemic variation from the frequency of use of each rule.

  • ICASSP - Extraction of phonemic variation rules in Continuous Speech spoken by multiple speakers
    ICASSP '87. IEEE International Conference on Acoustics Speech and Signal Processing, 1
    Co-Authors: S. Kimura, Y. Nara
    Abstract:

    This paper describes an interactive extraction of phonemic variation rules in Continuous Speech spoken by multiple speakers. To realize a Continuous Speech recognizer, we must first develop a highly accurate phoneme recognizer. The major problem related to phoneme recognizers is the phonemic variations in Continuous Speech. Our work focuses on the interactive analysis of phonemic variations in Continuous Speech and the extraction of the phonemic variation rules for many speakers. We extracted 317 rules related to 21 kinds of phonemic variation phenomena from 10,000 Japanese-language phrases spoken by 10 male speakers. With these rules, 97.6% of 36,000 Japanese-language phrases spoken by 36 test speakers (30 males and 6 females) were correctly segmented by our top-down phoneme segmentation system. Furthermore, a subset of the rules for each speaker was automatically obtained. On average, each subset contains 53.2% of the rules.

M. Dunham - One of the best experts on this subject based on the ideXlab platform.

  • ICASSP - BYBLOS: The BBN Continuous Speech recognition system
    ICASSP '87. IEEE International Conference on Acoustics Speech and Signal Processing, 1
    Co-Authors: Y. Chow, Owen Kimball, John Makhoul, M. Dunham, M. Krasner, G Kubala, Patti Price, S Roucos, R Schwartz
    Abstract:

    In this paper, we describe BYBLOS, the BBN Continuous Speech recognition system. The system, designed for large vocabulary applications, integrates acoustic, phonetic, lexical, and linguistic knowledge sources to achieve high recognition performance. The basic approach, as described in previous papers [1, 2], makes extensive use of robust context-dependent models of phonetic coarticulation using Hidden Markov Models (HMM). We describe the components of the BYBLOS system, including: signal processing frontend, dictionary, phonetic model training system, word model generator, grammar and decoder. In recognition experiments, we demonstrate consistently high word recognition performance on Continuous Speech across: speakers, task domains, and grammars of varying complexity. In speaker-dependent mode, where 15 minutes of Speech is required for training to a speaker, 98.5% word accuracy has been achieved in Continuous Speech for a 350-word task, using grammars with perplexity ranging from 30 to 60. With only 15 seconds of training Speech we demonstrate performance of 97% using a grammar.

  • ICASSP - A stochastic segment model for phoneme-based Continuous Speech recognition
    ICASSP '87. IEEE International Conference on Acoustics Speech and Signal Processing, 1
    Co-Authors: S Roucos, M. Dunham
    Abstract:

    Developing accurate and robust phonetic models for the different Speech sounds is a major challenge for high performance Continuous Speech recognition. In this paper, we introduce a new approach, called the stochastic segment model, for modelling a variable-length phonetic segment X, an L-long sequence of feature vectors. The stochastic segment model consists of 1) time-warping the variable-length segment X into a fixed-length segment Y called a resampled segment, and 2) a joint density function of the parameters of the resampled segment Y, which in this work is assumed Gaussian. In this paper, we describe the stochastic segment model, the recognition algorithm, and the iterative training algorithm for estimating segment models from Continuous Speech. For speaker-dependent Continuous Speech recognition, the segment model reduces the word error rate by one third over a hidden Markov phonetic model.

Osamu Kakusho - One of the best experts on this subject based on the ideXlab platform.

  • ICASSP - A Continuous Speech recognition system based on knowledge engineering techniques
    ICASSP '86. IEEE International Conference on Acoustics Speech and Signal Processing, 1
    Co-Authors: Riichiro Mizoguchi, Katsuhiko Tsujino, Osamu Kakusho
    Abstract:

    The main objective of our research is to construct a Continuous Speech recognition system by using knowledge engineering techniques. The system simulates the behavior of a human expert who can recognize Continuous Speech by inspecting the trajectories of feature parameters such as formant frequencies. The expertise is embedded in the form of production rules. In the current implementation, we have 114 rules which are being updated. The introduction of production system to knowledge representation enables to cope with large amount of heuristics of Speech recognition. Recognition rate of 85% was obtained for Continuous Speech of 30 second long uttered by three male adults. An environment for rule-base construction is also being developed.