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.
Voice Disorder Analysis: a Transformer-based Approach / Koudounas, Alkis; Ciravegna, Gabriele; Fantini, Marco; Crosetti, Erika; Succo, Giovanni; Cerquitelli, Tania; Baralis, Elena. - (2024), pp. 3040-3044. (Intervento presentato al convegno Interspeech 2024 tenutosi a Kos (GR) nel 1-5 September, 2024) [10.21437/interspeech.2024-1122].
Voice Disorder Analysis: a Transformer-based Approach
Koudounas, Alkis;Ciravegna, Gabriele;Cerquitelli, Tania;Baralis, Elena
2024
Abstract
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.File | Dimensione | Formato | |
---|---|---|---|
koudounas24_interspeech.pdf
accesso aperto
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
PUBBLICO - Tutti i diritti riservati
Dimensione
626.68 kB
Formato
Adobe PDF
|
626.68 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2992883