Data Augmentation with ChatGPT for Text to Sign Language Gloss Translation
Yazarlar (1)
Dr. Öğr. Üyesi Senem TANBERK Huawei R&D İstanbul, Türkiye
| Bildiri Türü | Tebliğ/Bildiri | Bildiri Dili | İngilizce |
| Bildiri Alt Türü | Tam Metin Olarak Yayınlanan Tebliğ (Uluslararası Kongre/Sempozyum) | ||
| Bildiri Niteliği | Alanında Hakemli Uluslararası Kongre/Sempozyum | ||
| Kongre Adı | 2024 International Conference on Emerging eLearning Technologies and Applications (ICETA) | ||
| Kongre Tarihi | 24-10-2024 / 24-10-2024 | ||
| Basıldığı Ülke | Türkiye | Basıldığı Şehir | |
| Bildiri Linki | https://ieeexplore.ieee.org/abstract/document/10850860/ | ||
| UAK Araştırma Alanları |
Eğitim Bilimleri
|
||
| Özet |
| In recent years, despite data limitations, sign language studies have attracted researchers' attention. For translating between spoken language and sign language gloss, which is considered a machine translation problem, data augmentation and state-of-the-art techniques have been proposed to overcome poor performance caused by data scarcity. This study focused on data augmentation for American Sign Language and British Sign Language using custom instructions ChatGPT4.0. Two different data selection methods were proposed to guide ChatGPT-4.0 in data generation. After data augmentation, the M2M100-418M model was fine-tuned on the original and augmented dataset. Experiments on the test data show that our method improves text-to-gloss translation performance. |
| Anahtar Kelimeler |
