In most approaches to speech recognition, the speech signals are segmented using constant-time segmentation, for example into 25 ms blocks. Constant segmentation risks losing info...
This paper presents a system which analyzes an in'put text syntactically and morphologically and converts the text from the graphemic to the phonetic :representation (or vice...
Most hypermedia systems emphasize the integration of graphics, images, video, and audio into a traditional hypertext framework. The hyperspeech system described in this paper, a s...
This paper is focused on the evaluation of listening test that was realized with a view to objectively annotate expressive speech recordings and further develop a limited domain ex...
Polish is a synthetic language with a high morpheme-perword ratio. It makes use of a high degree of inflection leading to high out-of-vocabulary (OOV) rates, and high Language Mo...
M. Ali Basha Shaik, Amr El-Desoky Mousa, Ralf Schl...