Update of acoustic and language models is vital to maintain performance of automatic speech recognition (ASR) systems. To alleviate efforts for updating models, we propose a "...
Yuya Akita, Masato Mimura, Graham Neubig, Tatsuya ...
Abstract. We propose a new class of tree automata, called tree automata with normalization (TAN). This framework extends equational tree automata, and improved the results of them:...
Many important applications, such as those using sparse data structures, have memory reference patterns that are unknown at compile-time. Prior work has developed runtime reorderi...
Michelle Mills Strout, Larry Carter, Jeanne Ferran...
— This paper demonstrates how two different sets of powerful domain specific language features can be specified and deployed as composable language extensions. These extensions...
This paper presents a Component-based Framework Development Process, of the Cardiology Domain. The Framework, called FrameCardio, was developed in 4 steps: 1Problem Domain Definit...