A comparison of three speaker-intrinsic vowel formant frequency normalization algorithms for sociophonetics

Anne Fabricius, Dominic Watt, Daniel Ezra Johnson

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

    Resumé

    This paper evaluates a speaker-intrinsic vowel formant frequency normalization algorithm initially proposed in Watt & Fabricius (2002). We compare how well this routine, known as the S-centroid procedure, performs as a sociophonetic research tool in three ways: reducing variance in area ratios of vowel spaces (thus attempting to equalize vowel space areas); improving overlap/co-extensiveness of vowel polygons; and reproducing two-dimensional vowel configurational relationships within the vowel space, compared to formant data in raw Hertz. The study uses existing datasets of vowel formant data from RP and Aberdeen English (northeast Scotland). We conclude that, for the data examined here, the S-centroid W&F procedures performs at least as well as the two most recognized speaker-intrinsic, vowel-extrinsic, formant-intrinsic normalization methods, Lobanov's (1971) z-score procedure and Nearey's (1977/8) individual log-mean procedure (CLIHi4 in Adank 2003, CLIHi2 as tested here), and in some test cases better than Nearey1.
    OriginalsprogEngelsk
    TidsskriftLanguage Variation and Change
    Vol/bind21
    Udgave nummer3
    Sider (fra-til)413-435
    ISSN0954-3945
    DOI
    StatusUdgivet - 2009

    Citer dette

    @article{d9f3e7400e2611debdfd000ea68e967b,
    title = "A comparison of three speaker-intrinsic vowel formant frequency normalization algorithms for sociophonetics",
    abstract = "This paper evaluates a speaker-intrinsic vowel formant frequency normalization algorithm initially proposed in Watt & Fabricius (2002). We compare how well this routine, known as the S-centroid procedure, performs as a sociophonetic research tool in three ways: reducing variance in area ratios of vowel spaces (thus attempting to equalize vowel space areas); improving overlap/co-extensiveness of vowel polygons; and reproducing two-dimensional vowel configurational relationships within the vowel space, compared to formant data in raw Hertz. The study uses existing datasets of vowel formant data from RP and Aberdeen English (northeast Scotland). We conclude that, for the data examined here, the S-centroid W&F procedures performs at least as well as the two most recognized speaker-intrinsic, vowel-extrinsic, formant-intrinsic normalization methods, Lobanov's (1971) z-score procedure and Nearey's (1977/8) individual log-mean procedure (CLIHi4 in Adank 2003, CLIHi2 as tested here), and in some test cases better than Nearey1.",
    author = "Anne Fabricius and Dominic Watt and Johnson, {Daniel Ezra}",
    year = "2009",
    doi = "10.1017/S0954394509990160",
    language = "English",
    volume = "21",
    pages = "413--435",
    journal = "Language Variation and Change",
    issn = "0954-3945",
    publisher = "Cambridge University Press",
    number = "3",

    }

    A comparison of three speaker-intrinsic vowel formant frequency normalization algorithms for sociophonetics. / Fabricius, Anne; Watt, Dominic; Johnson, Daniel Ezra.

    I: Language Variation and Change, Bind 21, Nr. 3, 2009, s. 413-435.

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

    TY - JOUR

    T1 - A comparison of three speaker-intrinsic vowel formant frequency normalization algorithms for sociophonetics

    AU - Fabricius, Anne

    AU - Watt, Dominic

    AU - Johnson, Daniel Ezra

    PY - 2009

    Y1 - 2009

    N2 - This paper evaluates a speaker-intrinsic vowel formant frequency normalization algorithm initially proposed in Watt & Fabricius (2002). We compare how well this routine, known as the S-centroid procedure, performs as a sociophonetic research tool in three ways: reducing variance in area ratios of vowel spaces (thus attempting to equalize vowel space areas); improving overlap/co-extensiveness of vowel polygons; and reproducing two-dimensional vowel configurational relationships within the vowel space, compared to formant data in raw Hertz. The study uses existing datasets of vowel formant data from RP and Aberdeen English (northeast Scotland). We conclude that, for the data examined here, the S-centroid W&F procedures performs at least as well as the two most recognized speaker-intrinsic, vowel-extrinsic, formant-intrinsic normalization methods, Lobanov's (1971) z-score procedure and Nearey's (1977/8) individual log-mean procedure (CLIHi4 in Adank 2003, CLIHi2 as tested here), and in some test cases better than Nearey1.

    AB - This paper evaluates a speaker-intrinsic vowel formant frequency normalization algorithm initially proposed in Watt & Fabricius (2002). We compare how well this routine, known as the S-centroid procedure, performs as a sociophonetic research tool in three ways: reducing variance in area ratios of vowel spaces (thus attempting to equalize vowel space areas); improving overlap/co-extensiveness of vowel polygons; and reproducing two-dimensional vowel configurational relationships within the vowel space, compared to formant data in raw Hertz. The study uses existing datasets of vowel formant data from RP and Aberdeen English (northeast Scotland). We conclude that, for the data examined here, the S-centroid W&F procedures performs at least as well as the two most recognized speaker-intrinsic, vowel-extrinsic, formant-intrinsic normalization methods, Lobanov's (1971) z-score procedure and Nearey's (1977/8) individual log-mean procedure (CLIHi4 in Adank 2003, CLIHi2 as tested here), and in some test cases better than Nearey1.

    U2 - 10.1017/S0954394509990160

    DO - 10.1017/S0954394509990160

    M3 - Journal article

    VL - 21

    SP - 413

    EP - 435

    JO - Language Variation and Change

    JF - Language Variation and Change

    SN - 0954-3945

    IS - 3

    ER -