Related Experiment Video
Updated: May 22, 2026
05:48
Memorization-Based Training and Testing Paradigm for Robust Vocal Identity Recognition in Expressive Speech Using Event-Related Potentials Analysis
Published on: August 9, 2024
Inter-speaker speech variability assessment using statistical deformable models from 3.0 tesla magnetic resonance
Maria J M Vasconcelos1, Sandra M R Ventura, Diamantino R S Freitas
1Faculty of Engineering, University of Porto/Institute of Mechanical Engineering and Industrial Management, Porto, Portugal.
Summary
This study introduces an automatic method using statistical deformable models to analyze vocal tract shape from 3.0 Tesla MRI scans. The technique accurately extracts vocal tract geometry and articulatory movements for speech production research.
Area of Science:
- Medical Imaging
- Speech Science
- Computational Anatomy
Background:
- Advancements in 3.0 Tesla magnetic resonance (MR) imaging enhance vocal tract analysis.
- Understanding vocal tract morphology and dynamics is crucial for speech production research.
Purpose of the Study:
- To automatically analyze vocal tract shape from 3.0 Tesla MR images using statistical deformable models.
- To evaluate the accuracy of automatic segmentation for vocal tract identification in new MR images.
- To characterize and reconstruct vocal tract shape during European Portuguese sound articulation.
Main Methods:
- Application of statistical deformable models, specifically active and appearance shape models.
- Utilizing 3.0 Tesla MR images for detailed vocal tract imaging.
- Automatic segmentation and shape extraction techniques.
Main Results:
- Demonstrated the effectiveness of statistical deformable models for automatic vocal tract analysis in 3.0 Tesla MR images.
- Successfully extracted vocal tract shape and assessed articulatory movements.
- Validated the automatic segmentation accuracy for vocal tract identification.
Conclusions:
- The automatic analysis of 3.0 Tesla MR images using deformable models is adequate and advantageous for extracting vocal tract shape and assessing articulatory movements.
- This method provides valuable insights into speech production, aiding in the study of articulatory disorders and the development of speech synthesizers.
