Keras implementation of Csapó T.G. & Sepúlveda-Sepúlveda A., ,,Ultrasound Tongue Image Synthesis for Acoustic-to-Articulatory Inversion using Convolutional and Recurrent Neural Networks'', submitted to Infocommunications Journal, 2022.
private, but can be sent upon request to the first author.
Trained models, predicted videos : https://simonyi-my.sharepoint.com/:f:/g/personal/csapszi_sch_bme_hu/EgIc-DOVUmJBjJ-DSTTgsHEBxVroMwU4qNC1bbyfI6TUVQ?e=C42y2Z
Ultrasound video samples: http://smartlab.tmit.bme.hu/UTI-inversion
Requirements
- Python 3.6.9
- ... see requirements.txt