DOAJ Open Access 2025

Can speech foundation models effectively identify languages in low-resource multilingual aging populations?

Aditya Kommineni Rajat Hebbar Sarah Petrosyan Pranali Khobragade Sudarsana Kadiri +3 lainnya

Lihat Sumber DOI

Abstrak

Speech foundation models (SFMs) achieve state-of-the-art results in many tasks, but their performance on elderly, multilingual speech remains underexplored. In this work, we investigate SFMs' ability to analyze multilingual speech from older adults using spoken language identification as a proxy task. We propose three key qualities for foundation models to serve multilingual aging populations: robustness to input duration, invariance to speaker demographics, and few-shot transferability in low-resource settings. Zero-shot evaluation indicates a noticeable performance drop for shorter inputs. We find that native speakers' speech consistently outperforms non-native speech across languages. Few-shot learning indicates better transferability in larger models.

Topik & Kata Kunci

Acoustics. Sound

Penulis (8)

Aditya Kommineni

Rajat Hebbar

Sarah Petrosyan

Pranali Khobragade

Sudarsana Kadiri

Miguel Arce Rentería

Jinkook Lee

Shrikanth Narayanan

Format Sitasi

APA MLA BibTeX

Kommineni, A., Hebbar, R., Petrosyan, S., Khobragade, P., Kadiri, S., Rentería, M.A. et al. (2025). Can speech foundation models effectively identify languages in low-resource multilingual aging populations?. https://doi.org/10.1121/10.0039265

Akses Cepat

Lihat di Sumber doi.org/10.1121/10.0039265

Informasi Jurnal

Tahun Terbit: 2025
Sumber Database: DOAJ
DOI: 10.1121/10.0039265
Akses: Open Access ✓