New
- Several available research positions in the areas of speech recognition and speech synthesis.
- Natural Speech Technology (NST): EPSRC Programme Grant with Cambridge and Sheffield (2011--2016)
- Ultrax: EPSRC Healthcare Partnerships grant on real-time tongue tracking for speech therapy with Queen Margaret University Speech Science Research Centre and Articulate Instruments (2011--2014)
- Cisco University Research Program Fund Grant on Speech Recognition (2010-2011)
- HSCMA 2011: Third Joint Workshop on Hands-free Speech Communication and Microphone Arrays, Edinburgh, 30 May - 1 June 2011
- Ravichander Vipperla successful PhD viva: Automatic speech recognition for ageing voices
- Joao Cabral awarded PhD: HMM-based Speech Synthesis using an Acoustic Glottal Source Model
- L. Lu, A. Ghoshal and S. Renals (2011). Regularized Subspace Gaussian Mixture Models for Speech Recognition, IEEE Signal Processing Letters, in press.
- J. Cabral, S. Renals, J. Yamagishi and K. Richmond (2011). HMM-based speech synthesiser using the LF-model of the glottal source, ICASSP-2011.
- J. Kilgour, J. Carletta and S. Renals (2011). The Ambient Spotlight: Personal meeting capture with a microphone array, HSCMA-2011.
- S. Huang and S. Renals (2010). Power Law Discounting for N-Gram Language Models, ICASSP-2010. (ICASSP award for best student paper in speech technology!).
- Co-editor-in-chief of ACM Transactions on Speech and Language Processing (2010-2013)
- Associate editor of IEEE Transaction on Audio, Speech, and Language Processing (2010-2013)
- Member of ISCA Advisory Council (2011-2014)
Research
I'm interested in understanding human communication using machine learning and statistical models, and constructing systems that can recognize and interpret communication scenes. My research career is grounded in speech processing, and our approaches start from the signals.
Speech Recognition and Synthesis
How can we improve conversational speech recognition? We are looking at better acoustic models that are discriminatively trained, that are better adapted or normalised to new domains or speakers, or that use improved spectral representations. Years of research have proven that it is difficult to improve upon appropriately smoothed n-grams for speech recognition - but we believe that non-parametric Bayesian models have some new things to offer. More recently, I've become interested in models, such as trajectory HMMs, that may be used for both recognition and synthesis. Current research students in speech recognition and synthesis include Liang Lu (Subspace GMM acoustic models) and Erich Zwyssig (microphone array based speaker diarization). Researchers working with me on speech recognition include Peter Bell, Fergus McInnes, and Mike Lincoln. In speech synthesis, these days I just try to keep up with the great work of Simon King, Junichi Yamagishi, and their colleagues, as well as working with Korin Richmond on tongue modelling.Read more...
Multimodal Interaction
Human communication is factored across more than one modality. The analysis and interpretation of multimodal interaction presents a number of challenges, ranging from ways to model multiple asynchronous streams of data to the construction of systems that can interpret aspects of multiparty meetings. A lot of this work is about augmenting communication in meetings (in the AMI and AMIDA Integrated Projects); we are also interested in the development of systems for home care. Current students in multimodal interaction include Karl Isaac, and I work with Maria Wolters, Jonathan Kilgour, and Jean Carletta in these areas. And I try to keep up with Hiroshi Shimodaira's work on synthesising conversational agents and social signals.Read more...
Projects
I'm principal investigator of the following projects: Natural Speech Technology, Ultrax, MultiMemo Home, and a Cisco University Research Program Fund Grant on Speech Recogniiton. I'm a co-investigator on SSPNet, SCALE, LISTA, and a UKIERI project on the Study of Source Features for Speech Synthesis and Speaker Recognition. Previous projects include the AMI and AMIDA Integrated Projects.Opportunities
Several research positions are available in the areas of speech recognition and speech synthesis on the NST project. We are looking for excellent research students: see the page about PhD opportunities at CSTR. I am not looking for visiting interns for the foreseeable future.Teaching
In 2010-2011 I taught Informatics 2B (jointly with Kyriakos Kalorkoti) and Automatic Speech Recognition (jointly with Hiroshi Shimodaira). I'm on sabbatical and not teaching in 2011-2012.