The authors present an estimation of fatigue level within individual operators using voice analysis. One advantages of voice analysis is its utilization of already existing operator communication system hardware. From the driver viewpoint it's an unobtrusive, non-interfering, secondary task. The expected fatigue induced speech changes refer to the voice categories of intensity, rhythm, pause patterns, intonation, speech rate, articulation, and speech quality. Due to interindividual differences in speech pattern the authors recorded speaker dependent baselines under alert conditions. Furthermore, sophisticated classification tools (e.g. Support Vector Machine, Multi-Layer Perception) were applied to distinguish these different fatigue clusters. To validate the voice analysis predetermined speech samples gained from a car simulator based sleep deprivation study (N=35; 20.00-04.00 a.m.) are used. Using standard acoustic feature computation procedures the authors selected 1069 features and feed them into seven machine learning methods. After each combining the output of each single classifier the authors yield a recognition rate of 79.4 % in classifying slightfrom strong fatigue. Section 2 introduces the cognitive-physiological mediator-model of fatigue induced speech changes. In section 3 the procedure of computing acoustic features is explained. Section 4 describes the design of the sleep deprivation study used for building a fatigue speaker database. Having provided the results of the fatigue detection in section 5, the paper closes with a conclusion and a discussion of the future work in section 6.


    Access

    Access via TIB

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Acoustic sleepiness analysis - using speech input from driver assistance systems for the phonetic detection of critical driver states


    Additional title:

    Akustische Schläfrigkeitsanalyse - die Verwendung von Spracheingabe aus Fahrerassistenzsystemen für die phonetische Erkennung von kritischen Fahrerzuständen


    Contributors:
    Krajewski, J. (author) / Sommer, D. (author) / Schupp, T. (author) / Golz, M. (author)


    Publication date :

    2009


    Size :

    9 Seiten, 4 Bilder, 1 Tabelle, 14 Quellen




    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English




    Driver Sleepiness

    Horne, J. / Reyner, L. | British Library Conference Proceedings | 1997


    DRIVER SLEEPINESS DETECTION METHOD

    European Patent Office | 2019

    Free access

    DRIVER SLEEPINESS DETECTION METHOD

    LEE SU WOONG / KOH SUK GYU / LEE JONG IL et al. | European Patent Office | 2019

    Free access

    DRIVER SLEEPINESS

    Horne, J. / Reyner, L. | British Library Conference Proceedings | 1997


    Detecting driver sleepiness

    Sandberg, David | TIBKAT | 2011