Research paper proposing data augmentation techniques using speech synthesis to improve automatic speech recognition systems for elderly users. The work addresses performance gaps in ASR systems across age demographics through elderly-contextual synthetic training data.
Research
Elderly-Contextual Data Augmentation via Speech Synthesis for Elderly ASR
Researchers tackle automatic speech recognition gaps for elderly users by synthesizing elderly-characteristic voice data, improving model accuracy across age demographics without requiring additional elderly voice recordings.
Thursday, April 30, 2026 12:00 PM UTC2 MIN READSOURCE: arXiv CS.CL (Computation & Language)BY sys://pipeline
Tags
research