Modeling Lexical Tones for Speaker Discrimination

Bibliographic Data

ID6440693
AuthorsRicky Kw Chan (0000-0003-4977-8406, Speech, Language and Cognition Laboratory, School of English, The University of Hong Kong, Hong Kong, corresponding author), Bruce Xiao Wang (0000-0003-3564-2911, Hong Kong Polytechnic University)
Year2024
Volume68
Issue1
Pages229-243
Publication date2024-07-27
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueLanguage and Speech (JOURNAL)
Journal identifiersISSN: 0023-8309 • E-ISSN: 1756-6053
PublisherSAGE Publishing (PUBLISHER • US)
DOI10.1177/00238309241261702
PMID39066631
OpenAlexW4401051652
LanguageEN
References cited30

Fundamental frequency (F0) has been widely studied and used in the context of speaker discrimination and forensic voice comparison casework, but most previous studies focused on long-term F0 statistics. Lexical tone, the linguistically structured and dynamic aspects of F0, has received much less research attention. A main methodological issue lies on how tonal F0 should be parameterized for the best speaker discrimination performance. This paper compares the speaker discriminatory performance of three approaches with lexical tone modeling: discrete cosine transform (DCT), polynomial curve fitting, and quantitative target approximation (qTA). Results show that using parameters based on DCT and polynomials led to similarly promising performance, whereas those based on qTA generally yielded relatively poor performance. Implications modeling surface tonal F0 and the underlying articulatory processes for speaker discrimination are discussed

Context (archaeology · Discrete cosine transform · Linguistics · Speaker recognition · Speech recognition · Tone (literature · Artificial Intelligence · Computer Science · Phonetics and Phonology Research · Speech and Audio Processing · Speech Recognition and Synthesis

Tools

Open DOI