Abstract

In the past, research on speech respiration and articulation has mostly been conducted independently of each other. Due to the limited availability of methods for simultaneously measuring respiration and articulation, the coordination of these two subsystems to encode linguistic structures is yet to be fully examined. In this study, we present a protocol for performing co-registration of real-time vocal tract MRI and respiratory plethysmography for speech research. The procedures for data acquisition and analysis are first presented. Then, the preliminary results from two sample speech production experiments are reported. The first experiment is on the timing between articulation and respiration, and the second one examines articulatory and respiratory aspects of linguistic focus encoding. The implications of the method for speech research are discussed.

Date

October 9, 2025

Authors

Yubin Zhang, Prakash Kumar, Ye Tian, Ziwei Zhao, Xuan Shi, Kevin Huang, Kevin Lee, Haley Hsu, Shrikanth Narayanan, Krishna Nayak, Louis Goldstein

Conference

Proc. Interspeech 2025

Pages

983-987