Speaker-Dependent Feature Generalizability for the Detection of Alcohol Intoxication

Authors

  • Arian Shamei UBC
  • Xinglei Liu
  • Rima Seiilova

Keywords:

Acoustic automated detection, acoustic analysis, Artificial intelligence, Acoustic speech measures

Abstract

Impairments to speech motor control from alcohol intoxication are variable across individuals, making speaker-dependent approaches ideal for speech-based intoxication detection. Here we evaluated whether individual acoustic features have high generalizability across speaker-dependent models. We selected 97 speakers (54 male, 43 female) from the Alcohol Language Corpus  who had sufficient sober and intoxicated (>0.8% blood-alcohol concentration) recordings for speaker-dependent modeling. We extracted 17 features from each audio file, then fitted these features to speaker-dependent random forest models and evaluated feature importance using mean decrease in Gini impurity (GI). Results show that across all speakers, consonant-based features tend to have stronger generalizability than vowel-based features, with spectral skewness and kurtosis being the most generalizable (GI: .11 and .09), and vowel duration and F2 being the least generalizable (GI: .04 and .03).

Additional Files

Published

2024-05-12

How to Cite

1.
Shamei A, Liu X, Seiilova R. Speaker-Dependent Feature Generalizability for the Detection of Alcohol Intoxication. Canadian Acoustics [Internet]. 2024 May 12 [cited 2024 Oct. 7];52(1). Available from: https://jcaa.caa-aca.ca/index.php/jcaa/article/view/4197

Issue

Vol. 52 No. 1 (2024): Proceedings of Acoustics Week in Canada 2024 Conference joint with 186th Meeting of the Acoustical Society of America

Section

Proceedings of the Acoustics Week in Canada

License

Author Licensing Addendum

This Licensing Addendum ("Addendum") is entered into between the undersigned Author(s) and Canadian Acoustics journal published by the Canadian Acoustical Association (hereinafter referred to as the "Publisher"). The Author(s) and the Publisher agree as follows:

  1. Retained Rights: The Author(s) retain(s) the following rights:

    • The right to reproduce, distribute, and publicly display the Work on the Author's personal website or the website of the Author's institution.
    • The right to use the Work in the Author's teaching activities and presentations.
    • The right to include the Work in a compilation for the Author's personal use, not for sale.

  2. Grant of License: The Author(s) grant(s) to the Publisher a worldwide exclusive license to publish, reproduce, distribute, and display the Work in Canadian Acoustics and any other formats and media deemed appropriate by the Publisher.

  3. Attribution: The Publisher agrees to include proper attribution to the Author(s) in all publications and reproductions of the Work.

  4. No Conflict: This Addendum is intended to be in harmony with, and not in conflict with, the terms and conditions of the original agreement entered into between the Author(s) and the Publisher.

  5. Copyright Clause: Copyright on articles is held by the Author(s). The corresponding Author has the right to grant on behalf of all Authors and does grant on behalf of all Authors, a worldwide exclusive license to the Publisher and its licensees in perpetuity, in all forms, formats, and media (whether known now or created in the future), including but not limited to the rights to publish, reproduce, distribute, display, store, translate, create adaptations, reprints, include within collections, and create summaries, extracts, and/or abstracts of the Contribution.

Most read articles by the same author(s)