Introduction to the Issue on Automatic Assessment of Health Disorders Based on Voice, Speech, and Language Processing

Abstract

This special issue is expected to contribute to make this field more open and visible to speech and signal processing experts, presenting the state of the art and the potentiality of the current speech and language technologies to contribute to solve important challenges that were partially hidden to the scientific community. Moreover, it is expected to contribute to stimulate the multi and interdisciplinary work that has evolved the speech technologies from its very early stages, but with a turn to the biomedical field. The emphases are on both basic and applied research related to the monitoring of voice and speech production status, as well as in the clinical evaluation of new developments. From these papers, we hope that the interested reader will find useful suggestions and further stimulation to carry on research in this field. This special issue is expected to elucidate not only the technical, but also some of the clinical difficulties inherent to the field, providing a forum for sharing thoughts on how to overcome them. Although there has been significant progress in the field, there are many future challenges, including the interpretation of learned models, adversarial examples, the lack of reproducibility due to the non-existence of open corpora, the existence of comorbidities affecting the different disorders, and problems associated with data-poor domains.