Speech recognition of a named entity

Tatsuhiko Tomita, Yoshiyuki Okimoto, Hirofumi Yamamoto, Yoshinori Sagisaka

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    2 Citations (Scopus)

    Abstract

    A hierarchical language model is newly applied to identify a named entity consisting of multiple word sequences for continuous speech recognition. By redesigning an out-of-vocabulary model of a single word using phonotactic constraints for a named entity, a hierarchical model is composed harmoniously with conventional word and word-class N-grams. Continuous speech recognition experiments aiming at movie-title identification showed the effectiveness of this modeling in the task of inquiries on these titles. These results ensure that the proposed hierarchical language modeling architecture is applicable to multiple word successions for speech recognition to cope with unregistered expressions and enables the mix use of different statistics harmoniously.

    Original languageEnglish
    Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    VolumeI
    ISBN (Print)0780388747, 9780780388741
    DOIs
    Publication statusPublished - 2005
    Event2005 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP '05 - Philadelphia, PA
    Duration: 2005 Mar 182005 Mar 23

    Other

    Other2005 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP '05
    CityPhiladelphia, PA
    Period05/3/1805/3/23

    Fingerprint

    speech recognition
    Speech recognition
    Continuous speech recognition
    Statistics
    statistics
    Experiments

    ASJC Scopus subject areas

    • Electrical and Electronic Engineering
    • Signal Processing
    • Acoustics and Ultrasonics

    Cite this

    Tomita, T., Okimoto, Y., Yamamoto, H., & Sagisaka, Y. (2005). Speech recognition of a named entity. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (Vol. I). [1415299] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2005.1415299

    Speech recognition of a named entity. / Tomita, Tatsuhiko; Okimoto, Yoshiyuki; Yamamoto, Hirofumi; Sagisaka, Yoshinori.

    ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. I Institute of Electrical and Electronics Engineers Inc., 2005. 1415299.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Tomita, T, Okimoto, Y, Yamamoto, H & Sagisaka, Y 2005, Speech recognition of a named entity. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. vol. I, 1415299, Institute of Electrical and Electronics Engineers Inc., 2005 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP '05, Philadelphia, PA, 05/3/18. https://doi.org/10.1109/ICASSP.2005.1415299
    Tomita T, Okimoto Y, Yamamoto H, Sagisaka Y. Speech recognition of a named entity. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. I. Institute of Electrical and Electronics Engineers Inc. 2005. 1415299 https://doi.org/10.1109/ICASSP.2005.1415299
    Tomita, Tatsuhiko ; Okimoto, Yoshiyuki ; Yamamoto, Hirofumi ; Sagisaka, Yoshinori. / Speech recognition of a named entity. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. I Institute of Electrical and Electronics Engineers Inc., 2005.
    @inproceedings{b0e7ddea548045ecb248e70e50c0b358,
    title = "Speech recognition of a named entity",
    abstract = "A hierarchical language model is newly applied to identify a named entity consisting of multiple word sequences for continuous speech recognition. By redesigning an out-of-vocabulary model of a single word using phonotactic constraints for a named entity, a hierarchical model is composed harmoniously with conventional word and word-class N-grams. Continuous speech recognition experiments aiming at movie-title identification showed the effectiveness of this modeling in the task of inquiries on these titles. These results ensure that the proposed hierarchical language modeling architecture is applicable to multiple word successions for speech recognition to cope with unregistered expressions and enables the mix use of different statistics harmoniously.",
    author = "Tatsuhiko Tomita and Yoshiyuki Okimoto and Hirofumi Yamamoto and Yoshinori Sagisaka",
    year = "2005",
    doi = "10.1109/ICASSP.2005.1415299",
    language = "English",
    isbn = "0780388747",
    volume = "I",
    booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",

    }

    TY - GEN

    T1 - Speech recognition of a named entity

    AU - Tomita, Tatsuhiko

    AU - Okimoto, Yoshiyuki

    AU - Yamamoto, Hirofumi

    AU - Sagisaka, Yoshinori

    PY - 2005

    Y1 - 2005

    N2 - A hierarchical language model is newly applied to identify a named entity consisting of multiple word sequences for continuous speech recognition. By redesigning an out-of-vocabulary model of a single word using phonotactic constraints for a named entity, a hierarchical model is composed harmoniously with conventional word and word-class N-grams. Continuous speech recognition experiments aiming at movie-title identification showed the effectiveness of this modeling in the task of inquiries on these titles. These results ensure that the proposed hierarchical language modeling architecture is applicable to multiple word successions for speech recognition to cope with unregistered expressions and enables the mix use of different statistics harmoniously.

    AB - A hierarchical language model is newly applied to identify a named entity consisting of multiple word sequences for continuous speech recognition. By redesigning an out-of-vocabulary model of a single word using phonotactic constraints for a named entity, a hierarchical model is composed harmoniously with conventional word and word-class N-grams. Continuous speech recognition experiments aiming at movie-title identification showed the effectiveness of this modeling in the task of inquiries on these titles. These results ensure that the proposed hierarchical language modeling architecture is applicable to multiple word successions for speech recognition to cope with unregistered expressions and enables the mix use of different statistics harmoniously.

    UR - http://www.scopus.com/inward/record.url?scp=33646768658&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=33646768658&partnerID=8YFLogxK

    U2 - 10.1109/ICASSP.2005.1415299

    DO - 10.1109/ICASSP.2005.1415299

    M3 - Conference contribution

    AN - SCOPUS:33646768658

    SN - 0780388747

    SN - 9780780388741

    VL - I

    BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -