Vocal tract area function estimation from three-dimensional (3D) volumetric dataset often involves complex and manual procedures such as oblique slice cutting and image segmentation. We introduce a semi-automatic method for estimating vocal tract area function from 3D Magnetic Resonance Imaging (MRI) datasets. The method was implemented on a custom MATLAB graphical user interface and computes the area function in a user-interactive way. The 3D MRI datasets were acquired with 1.25 mm isotropic resolution during 8-seconds sustained sound productions of vowels /IY/, /AA/, /UW/ by one male native speaker of American English at a 3 Tesla MRI scanner.
|Title of host publication||Proceedings of the Interspeech 2013 Workshop on Speech Production in Automatic Speech Recognition (SPASR)|
|Place of Publication||Chicago, Illinois|
|Publisher||Toyota Technological Institute at Chicago|
|Number of pages||4|
|Publication status||Published - 2013|
|Event||Workshop on Speech Production in Automatic Speech Recognition - Lyon, France|
Duration: 30 Aug 2013 → 30 Aug 2013
|Workshop||Workshop on Speech Production in Automatic Speech Recognition|
|Period||30/08/13 → 30/08/13|
- speech production
- magnetic resonance imaging
- image segmentation
- area function
- vocal tract shape
Kim, Y-C., Kim, J., Proctor, M., Toutios, A., Nayak, K., Lee, S., & Narayanan, S. (2013). Toward automatic vocal tract area function estimation from accelerated three-dimensional magnetic resonance imaging. In K. Livescu (Ed.), Proceedings of the Interspeech 2013 Workshop on Speech Production in Automatic Speech Recognition (SPASR) (pp. 1-4). Chicago, Illinois: Toyota Technological Institute at Chicago.