Africa-Centric Self-Supervised Pre-Training for Multilingual Speech Representation in a Sub-Saharan Context
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911847493402624 |
|---|---|
| author | Caubrière, Antoine Gauthier, Elodie |
| author_facet | Caubrière, Antoine Gauthier, Elodie |
| contents | We present the first self-supervised multilingual speech model trained exclusively on African speech. The model learned from nearly 60 000 hours of unlabeled speech segments in 21 languages and dialects spoken in sub-Saharan Africa. On the SSA subset of the FLEURS-102 dataset, our approach based on a HuBERT$_{base}$ (0.09B) architecture shows competitive results, for ASR downstream task, compared to the w2v-bert-51 (0.6B) pre-trained model proposed in the FLEURS benchmark, while being more efficient by using 7x less data and 6x less parameters. Furthermore, in the context of a LID downstream task, our approach outperforms FLEURS baselines accuracy by over 22\%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_02000 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Africa-Centric Self-Supervised Pre-Training for Multilingual Speech Representation in a Sub-Saharan Context Caubrière, Antoine Gauthier, Elodie Computation and Language Machine Learning Sound Audio and Speech Processing We present the first self-supervised multilingual speech model trained exclusively on African speech. The model learned from nearly 60 000 hours of unlabeled speech segments in 21 languages and dialects spoken in sub-Saharan Africa. On the SSA subset of the FLEURS-102 dataset, our approach based on a HuBERT$_{base}$ (0.09B) architecture shows competitive results, for ASR downstream task, compared to the w2v-bert-51 (0.6B) pre-trained model proposed in the FLEURS benchmark, while being more efficient by using 7x less data and 6x less parameters. Furthermore, in the context of a LID downstream task, our approach outperforms FLEURS baselines accuracy by over 22\%. |
| title | Africa-Centric Self-Supervised Pre-Training for Multilingual Speech Representation in a Sub-Saharan Context |
| topic | Computation and Language Machine Learning Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2404.02000 |