This paper provides evidence that the use of more unlabeled data in semi-supervised learning can improve the performance of Natural Language Processing (NLP) tasks, such as part-o...
Duration of phonemic segments provide important cues for distinguishing words in languages such as Arabic. Recently, we proposed a discriminatively estimated joint acoustic, durat...
In this study, we propose increasing discriminative power on the maximum a posteriori (MAP)-based mapping function estimation for acoustic model adaptation. Based on the effective...
This paper investigates syntactic and sub-lexical features in Turkish discriminative language models (DLMs). DLM is a featurebased language modeling approach. It reranks the ASR o...
Ebru Arisoy, Murat Saraclar, Brian Roark, Izhak Sh...
This paper proposes a novel approach to recognize object categories in point clouds. By quantizing 3D SURF local descriptors, computed on partial 3D shapes extracted from the poin...