Voice Disorder Analysis: a Transformer-based Approach
This work addresses the challenge of diagnosing voice disorders for patients, but it is incremental as it builds on existing transformer and data augmentation techniques.
The paper tackled the problem of non-invasive automated diagnosis of voice disorders by proposing a transformer-based approach that works directly on raw voice signals, using synthetic data generation and a Mixture of Expert ensemble to handle data shortage and diverse recording types, resulting in significant improvements over existing methods in detection and classification tasks.
Voice disorders are pathologies significantly affecting patient quality of life. However, non-invasive automated diagnosis of these pathologies is still under-explored, due to both a shortage of pathological voice data, and diversity of the recording types used for the diagnosis. This paper proposes a novel solution that adopts transformers directly working on raw voice signals and addresses data shortage through synthetic data generation and data augmentation. Further, we consider many recording types at the same time, such as sentence reading and sustained vowel emission, by employing a Mixture of Expert ensemble to align the predictions on different data types. The experimental results, obtained on both public and private datasets, show the effectiveness of our solution in the disorder detection and classification tasks and largely improve over existing approaches.