Development and evaluation of Japanese Lombard speech corpus

Tetsuji Ogawa, Takanobu Nishiura, Takeshi Yamada, Norihide Kitaoka, Tetsunori Kobayashi

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

    1 Citation (Scopus)

    Abstract

    We have developed a Japanese Lombard speech corpus suitable for a wide range of applications such as those for improving the performance of noisy speech and speaker recognition systems and analyzing acoustic changes introduced by the Lombard effect. The corpus contains clean speech data with neutral and Lombard talking styles and noisy speech data with the Lombard talking style. Its development was based on a sample size of 40 people speaking while being subjected to various types of noise at different sound pressure levels (SPLs). The evaluations were primarily targeted at speech and speaker recognition systems, but they also relate to the accuracy analysis of simulation-based assessments of noisy speech recognition systems. The impact of the Lombard effect on the recognition system performance is also discussed, as is the extent to which the impact varies across different types and SPLs of noise.

    Original languageEnglish
    Title of host publication40th International Congress and Exposition on Noise Control Engineering 2011, INTER-NOISE 2011
    Pages1366-1373
    Number of pages8
    Volume2
    Publication statusPublished - 2011
    Event40th International Congress and Exposition on Noise Control Engineering 2011, INTER-NOISE 2011 - Osaka, Japan
    Duration: 2011 Sep 42011 Sep 7

    Other

    Other40th International Congress and Exposition on Noise Control Engineering 2011, INTER-NOISE 2011
    CountryJapan
    CityOsaka
    Period11/9/411/9/7

    Fingerprint

    speech recognition
    talking
    sound pressure
    evaluation
    acoustics
    simulation

    Keywords

    • Lombard effect
    • Speaker recognition
    • Speech recognition

    ASJC Scopus subject areas

    • Acoustics and Ultrasonics

    Cite this

    Ogawa, T., Nishiura, T., Yamada, T., Kitaoka, N., & Kobayashi, T. (2011). Development and evaluation of Japanese Lombard speech corpus. In 40th International Congress and Exposition on Noise Control Engineering 2011, INTER-NOISE 2011 (Vol. 2, pp. 1366-1373)

    Development and evaluation of Japanese Lombard speech corpus. / Ogawa, Tetsuji; Nishiura, Takanobu; Yamada, Takeshi; Kitaoka, Norihide; Kobayashi, Tetsunori.

    40th International Congress and Exposition on Noise Control Engineering 2011, INTER-NOISE 2011. Vol. 2 2011. p. 1366-1373.

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

    Ogawa, T, Nishiura, T, Yamada, T, Kitaoka, N & Kobayashi, T 2011, Development and evaluation of Japanese Lombard speech corpus. in 40th International Congress and Exposition on Noise Control Engineering 2011, INTER-NOISE 2011. vol. 2, pp. 1366-1373, 40th International Congress and Exposition on Noise Control Engineering 2011, INTER-NOISE 2011, Osaka, Japan, 11/9/4.
    Ogawa T, Nishiura T, Yamada T, Kitaoka N, Kobayashi T. Development and evaluation of Japanese Lombard speech corpus. In 40th International Congress and Exposition on Noise Control Engineering 2011, INTER-NOISE 2011. Vol. 2. 2011. p. 1366-1373
    Ogawa, Tetsuji ; Nishiura, Takanobu ; Yamada, Takeshi ; Kitaoka, Norihide ; Kobayashi, Tetsunori. / Development and evaluation of Japanese Lombard speech corpus. 40th International Congress and Exposition on Noise Control Engineering 2011, INTER-NOISE 2011. Vol. 2 2011. pp. 1366-1373
    @inproceedings{81c4e2602f9b4fa19514409e131b8283,
    title = "Development and evaluation of Japanese Lombard speech corpus",
    abstract = "We have developed a Japanese Lombard speech corpus suitable for a wide range of applications such as those for improving the performance of noisy speech and speaker recognition systems and analyzing acoustic changes introduced by the Lombard effect. The corpus contains clean speech data with neutral and Lombard talking styles and noisy speech data with the Lombard talking style. Its development was based on a sample size of 40 people speaking while being subjected to various types of noise at different sound pressure levels (SPLs). The evaluations were primarily targeted at speech and speaker recognition systems, but they also relate to the accuracy analysis of simulation-based assessments of noisy speech recognition systems. The impact of the Lombard effect on the recognition system performance is also discussed, as is the extent to which the impact varies across different types and SPLs of noise.",
    keywords = "Lombard effect, Speaker recognition, Speech recognition",
    author = "Tetsuji Ogawa and Takanobu Nishiura and Takeshi Yamada and Norihide Kitaoka and Tetsunori Kobayashi",
    year = "2011",
    language = "English",
    isbn = "9781618392800",
    volume = "2",
    pages = "1366--1373",
    booktitle = "40th International Congress and Exposition on Noise Control Engineering 2011, INTER-NOISE 2011",

    }

    TY - GEN

    T1 - Development and evaluation of Japanese Lombard speech corpus

    AU - Ogawa, Tetsuji

    AU - Nishiura, Takanobu

    AU - Yamada, Takeshi

    AU - Kitaoka, Norihide

    AU - Kobayashi, Tetsunori

    PY - 2011

    Y1 - 2011

    N2 - We have developed a Japanese Lombard speech corpus suitable for a wide range of applications such as those for improving the performance of noisy speech and speaker recognition systems and analyzing acoustic changes introduced by the Lombard effect. The corpus contains clean speech data with neutral and Lombard talking styles and noisy speech data with the Lombard talking style. Its development was based on a sample size of 40 people speaking while being subjected to various types of noise at different sound pressure levels (SPLs). The evaluations were primarily targeted at speech and speaker recognition systems, but they also relate to the accuracy analysis of simulation-based assessments of noisy speech recognition systems. The impact of the Lombard effect on the recognition system performance is also discussed, as is the extent to which the impact varies across different types and SPLs of noise.

    AB - We have developed a Japanese Lombard speech corpus suitable for a wide range of applications such as those for improving the performance of noisy speech and speaker recognition systems and analyzing acoustic changes introduced by the Lombard effect. The corpus contains clean speech data with neutral and Lombard talking styles and noisy speech data with the Lombard talking style. Its development was based on a sample size of 40 people speaking while being subjected to various types of noise at different sound pressure levels (SPLs). The evaluations were primarily targeted at speech and speaker recognition systems, but they also relate to the accuracy analysis of simulation-based assessments of noisy speech recognition systems. The impact of the Lombard effect on the recognition system performance is also discussed, as is the extent to which the impact varies across different types and SPLs of noise.

    KW - Lombard effect

    KW - Speaker recognition

    KW - Speech recognition

    UR - http://www.scopus.com/inward/record.url?scp=84867977951&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84867977951&partnerID=8YFLogxK

    M3 - Conference contribution

    SN - 9781618392800

    VL - 2

    SP - 1366

    EP - 1373

    BT - 40th International Congress and Exposition on Noise Control Engineering 2011, INTER-NOISE 2011

    ER -