Making more of little data: Improving low-resource automatic speech recognition using data augmentation

M Bartelds, N San, B McDonnell, D Jurafsky… - ar** communities process restricted-access corpora for language revival efforts
N San, M Bartelds, T Ogunremi, A Mount… - arxiv preprint arxiv …, 2022 - arxiv.org
Many archival recordings of speech from endangered languages remain unannotated and
inaccessible to community members and language learning programs. One bottleneck is the …

From'snippet-lects' to doculects and dialects: Leveraging neural representations of speech for placing audio signals in a language landscape

S Guillaume, G Wisniewski, A Michaud - 2nd Annual Meeting of the …, 2023 - hal.science
XLSR-53, a multilingual model of speech, builds a vector representation from audio, which
allows for a range of computational treatments. The experiments reported here use this …

Unsupervised discovery of recurring spoken terms using diagonal patterns

P Sudhakar, K Sreenivasa Rao, P Mitra - International Conference on …, 2023 - Springer
Spoken term discovery is a challenging task when a lot of spoken content is generated
without annotation. The spoken term discovery task accomplished by pattern matching …

On the nature of discrete speech representations in multilingual self-supervised models

BM Abdullah, MM Shaik, D Klakow - Proceedings of the 5th …, 2023 - aclanthology.org
Self-supervision has emerged as an effective paradigm for learning representations of
spoken language from raw audio without explicit labels or transcriptions. Self-supervised …

Efficiency-oriented approaches for self-supervised speech representation learning

L Lugo, V Vielzeuf - International Journal of Speech Technology, 2024 - Springer
Self-supervised learning enables the training of large neural models without the need for
large, labeled datasets. It has been generating breakthroughs in several fields, including …