Speaker-normalized Acoustic Distance for Japanese Pitch Accent
Abstract
Cross-speaker comparison of Japanese pitch accents is unstable due to individual variation in F0 range, speech rate, and phonation. This work proposes a speaker-normalized F0-based distance using Dynamic Time Warping (DTW), demonstrating that interpretable distance design can match or exceed black-box representations for prosodic typology tasks.
Fields of Study & Subjects
Phonetics and Phonology Research
Speech Recognition and Synthesis
Music and Audio Processing
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 26 February 2026
- Language
- EN