Accent-robust speech recognition for English in low-resource settings using Manifold Mixup

Abstract We adapt Manifold Mixup theory for accent-robust end-to-end (E2E) Automatic speech recognition (ASR). Accent-variation between a source and target constitutes a domain-mismatch scenario. Manifold Mixup allows cross-domain robustness where a model trained on a source accent generalizes to ta...

Descripció completa

Dades bibliogràfiques

Autors principals:	Tirthankar Banerjee, V. Ramasubramanian
Format:	Artigo
Idioma:	Inglês
Publicat:	SpringerOpen 2025-11-01
Col·lecció:	EURASIP Journal on Audio, Speech, and Music Processing
Matèries:	Automated speech recognition Accented speech recognition Manifold mixup Cross accent Indic TIMIT Svarah
Accés en línia:	https://doi.org/10.1186/s13636-025-00435-0
Etiquetes:	Afegir etiqueta Sense etiquetes, Sigues el primer a etiquetar aquest registre!