A study on the lie detection of telephone voices using support vector machine
- Authors
- Park, H.; Kim, J.-B.; Bae, S.-G.; Kim, M.-S.
- Issue Date
- Jan-2019
- Publisher
- Springer Verlag
- Keywords
- Credit evaluation; Lie detection; Telephone voice; Voice analysis
- Citation
- Studies in Computational Intelligence, v.790, pp.89 - 99
- Journal Title
- Studies in Computational Intelligence
- Volume
- 790
- Start Page
- 89
- End Page
- 99
- URI
- http://scholarworks.bwise.kr/ssu/handle/2018.sw.ssu/31357
- DOI
- 10.1007/978-3-319-98367-7_8
- ISSN
- 1860-949X
- Abstract
- Human voices are one of the easiest ways to communicate information between humans. Voice characteristics may vary from person to person and include voice rate, genital form and function, pitch tone, language habits, and gender. Human voices are a key element of human communication. In the era of the Fourth Industrial Revolution, the voices are the main means of communication between people and people, between humans and machines, machines and machines. And for that reason, people are trying to communicate their intent clearly to others. In the process, language information and various additional information are included. Information such as emotional state, health status, reliability, presence of lies, changes due to alcohol, etc. These languages and non-linguistic information can be used as a device to assess the lie of telephone voices that appear as various parameters. Especially, it can be obtained by analyzing the relationship between the characteristics of the fundamental frequency (fundamental tone) of the vocal cords and the resonance frequency characteristics of vocal tracks. Previous studies have extracted parameters for false testimony of various telephone voices and conducted this study to evaluate whether a telephone voice is a lie. In this study, we proposed a judge to judge whether a lie is true by using a support vector machine. We propose a personal telephone truth discriminator. ? 2019, Springer Nature Switzerland AG.
- Files in This Item
-
Go to Link
- Appears in
Collections - College of Humanities > Department of English Language & Literature > 1. Journal Articles
- Graduate School of Software > Major in Software > 1. Journal Articles
- College of Information Technology > ETC > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.