We invite applications for a position as Research Associate in the Speech and Hearing (SpandH) research group at Sheffield University to work on research into novel algorithms with application to automatic speech recognition, and possibly to speaker identification or language recognition. The post holder will work in the context of several projects, most notably the EPSRC programme on Natural Speech  Technology (www.natural-speech-technology.org), and  the project DocuMeet (www.documeet.eu). These projects involve collaboration with outstanding partners from academia and industry on technological topics such as canonical acoustic modelling, deep neural networks, noise robustness or speaker adaptation. Our research focus is on technology that is capable of performing well in diverse natural situations. This includes research on media archive processing, meeting and interview recognition, as well as lecture processing. The challenges in these areas are
manifold: diverse backgrounds, conversational speech – potentially heavily accented, cross talk, non-speech noise, or complex acoustic environment. Our aim is to improve recognition results by development of fundamental algorithms and techniques in the context of large-scale, state-of-the-art recognition systems. Work in this position should significantly advance research on one or more of the above challenges.

Applicants should have a track record of work on speech technologies including speech recognition, and have had exposure to modern machine learning techniques. Ideally, such a track record is demonstrated by publications in international journals and conferences. Candidates are required to hold a PhD in the field; work on the project will require publication of results,