Peter Bell

Peter Bell holds the Personal Chair of Speech Technology at The University of Edinburgh's Department of Computer Science. His research focuses on speech technology, though specific details are not provided. He is affiliated with the university and can be contacted via email at peter.bell@ed.ac.uk. His professional profile is available at https://www.ed.ac.uk/Peter_Bell.html.

Peter Bell is a researcher at the University of Edinburgh specializing in Automatic speech recognition. His work focuses on improving speech recognition systems through advanced techniques such as dialect detection, multi-genre broadcast media recognition, and adaptation algorithms for neural network-based speech recognition. He has contributed to the development of tools like ALISA for speech segmentation and alignment. His research also explores the use of self-supervised models, sequence-to-sequence frameworks, and multi-level adaptive networks to enhance speech recognition accuracy, particularly for low-resource languages and disordered speech. Bell's work emphasizes the integration of acoustic and lexical features for robust speech processing.