Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 144 (4), EL290

ACT: An Automatic Centroid Tracking Tool for Analyzing Vocal Tract Actions in Real-Time Magnetic Resonance Imaging Speech Production Data

Affiliations

ACT: An Automatic Centroid Tracking Tool for Analyzing Vocal Tract Actions in Real-Time Magnetic Resonance Imaging Speech Production Data

Miran Oh et al. J Acoust Soc Am.

Abstract

Real-time magnetic resonance imaging (MRI) speech production data have expanded the understanding of vocal tract actions. This letter presents an Automatic Centroid Tracking tool, ACT, which obtains both spatial and temporal information characterizing multi-directional articulatory movement. ACT auto-segments an articulatory object composed of connected pixels in a real-time MRI video, by finding its intensity centroids over time and returns kinematic profiles including direction and magnitude information of the object. This letter discusses the utility of ACT, which outperforms other similar object tracking techniques, by demonstrating its successful online tracking of vertical larynx movement. ACT can be deployed generally for dynamic image processing and analysis.

Figures

Fig. 1.
Fig. 1.
(Color online) Example processing steps of object segmentation and centroid tracking in ACT (a user-selected seed is indicated by a yellow asterisk [*]).
Fig. 2.
Fig. 2.
(Color online) Sample visualization of ACT tracking of the vertical larynx movement during the production of /ɑɠɑ/ (ROI size: 12 [width] × 40 [height] (in millimeters); rtMRI IPA data available online at http://sail.usc.edu/span/rtmri_ipa).
Fig. 3.
Fig. 3.
(Color online) (a) Correlation between f0 and the corresponding vertical larynx (black dotted line: regression line, gray dots: values measured from all the voiced intervals) and f0 and the corresponding vertical larynx centroid values for tense (red dots) vs lax (blue dots) from a female speaker of Seoul Korean, (b) sample vertical larynx movement time functions in Hausa ejective /k'/ (red line) and implosive /ɓ/ (gray line), and (c) vertical larynx position at movement maximum in Hausa ejectives (/s', k', kw'/) and implosives (/ɓ, ɗ/) (box heights: interquartile range [IQR], white dots: means; gray dots: outliers; horizontal lines: medians; vertical line heights: intervals between minimum and maximum values within 1.5 × IQR).

Similar articles

See all similar articles

Publication types

Feedback