People and Projects
Out of date, will be updated soon (2019.10.31)
People
Current PhD students
- Siva Reddy Gangireddy (Neural network language models). Supported by uDialogue.
- Ahmed Ali (Multi-dialect Arabic speech recognition). Partly supported by SUMMA.
- Joachim Fainberg (Multi-domain speech recognition). Supported by Bloomberg.
- Ben Krause (Recurrent neural networks). Support by Principal's Career Development award.
- Ondrej Klejch (Multilingual broadcast speech recognition). Supported by SUMMA.
Researchers working with me
- Peter Bell (Acoustic modelling, speech recognition systems)
- Jean Carletta (Multimodal interaction, group interaction)
- Jonathan Kilgour (Multimodal systems)
- Catherine Lai (Multimodal language processing, prosody)
- Liang Lu (Acoustic modelling)
- Korin Richmond (Articulatory modelling, speech synthesis)
PhD graduates
- Gethin Williams (Knowing What You Don't Know: Roles for Confidence Measures in Automatics Speech Recognition), 1999
- Miguel Carreira Perpiñán (Continuous latent variable models for dimensionality reduction and sequential data reconstruction), 2001
- Costis Kompis (Automatic Voicemail Summarisation for Mobile Messaging), 2002
- Vincent Wan (Speaker Verification using Support Vector Machines), 2003
- Yasser Hifny (Conditional Random Fields for Continuous Speech Recognition), 2006
- Gabriel Murray (Using Speech-Specific Characteristics for Automatic Speech Summarization), 2008
- Alfred Dielmann (Dynamic Bayesian networks for meeting segmentation), 2009
- Giulia Garau (Combining spectral representations and speaker normalisation), 2009
- Le Zhang (Modelling Speech Dynamics with Trajectory-HMMs), 2009
- Heriberto Cuayáhuitl (Reinforcement learning of spoken dialogue strategies), 2009
- Songfang Huang (Hierarchical Bayesian Language Models for Multiparty Conversational Speech Recognition), 2010
- João Cabral (HMM-based Speech Synthesis using an Acoustic Glottal Source Model), 2010
- Ravi Chander Vipperla (Automatic speech recognition for ageing voices), 2011
- Erich Zwyssig (Speech processing using digital MEMS microphones), 2013
- Liang Lu (Subspace Gaussian Mixture Models for Automatic Speech Recognition), 2013
- Karl Isaac (The Intelligibility of Synthetic Speech in Noise and Reverberation), 2015.
- Pawel Swietojanski (Learning Representations for Speech Recognition using Artificial Neural Networks). Supported by Natural Speech Technology, 2016.
Projects
Current projects
- SUMMA (EC H2020 project, 3 years, 2016-2019)
- EPSRC Doctoral Training Centre in Data Science (8 years; 2014-2022)
- uDialogue [joint with Nagoya Institute of Technology] (JST CREST, 5 years, 2011-2016)
Some previous projects
- Natural Speech Technology [joint with Cambridge and Sheffield Universities] (EPSRC programme grant, 5 years, 2011-2016)
- EPSRC Imapact Acceleration project on Just-in-time speech recognition for TV news
- Ultrax [joint with QMU and Articulate Instruments] (EPSRC, 3.5 years, 2011-2014)
- InEvent (EC IST STREP, 3 years, 2011-2014)
- EU-Bridge (EC IST IP, 3 years, 2012-2015; PI: Philipp Koehn)
- SSPNet: Social Signal Processing Network (EC IST NoE, 5 years, 2009-2014)
- SCALE: Speech Communication with Adaptive Learning (Marie Curie ITN, 4 years, 2009-2013)
- MultiMemo Home [joint with Glasgow and QMU] (EPSRC, 3.5 years, 2009-2013)
- EMIME: Effective Multilingual Interaction in Mobile Environments (EC IST STREP, 3 years, 2008-2011, PI: Simon King)
- Edinburgh Speech Production Facility (EPSRC, 3 years, 2007-2010, PI: Alice Turk)
- AMIDA: Augmented Multiparty Interaction with Distance Access (EC IST Integrated Project. 3 years, 2006-2009)
- EdSST: Edinburgh Speech Science and Technology (Marie Curie Early Stage Training Site, 4 years, 2006-2009)
- Data-driven articulatory modelling (EPSRC, 3 years, 2006-2009)
- MATCH: Mobilising Advanced Technology for Care at Home (SFC SRDG project, 4 years, 2005-2009)
- AMI: Augmented Multiparty Interaction (EC IST Integrated Project. 3 years, 2004-2006)
- M4: MultiModal Meeting Manager (EC IST Programme, 3 years, 2002-2005)
- S3L: Statistical Summarization of Spoken Language (EPSRC, 3 years, 2002-2005)
- SWAG: Spoken Word Archive Group (EU/US Working Group supported by DELOS in Europe and NSF in the US, 2001-2003)
- Speech Summarization using Prosodic Information (EPSRC ROPA 2001-2002)
- SToBS: Structured Transcription of Broadcast Speech (EPSRC 1998-2001)
- The THISL Spoken Document Retrieval Project (ESPRIT Long Term Research Programme 1997-2000)
- SPRACH: Speech Recognition Algorithms for Connectionist Hybrids (ESPRIT Long Term Research Programme 1995-1998)
- Wernicke: A Neural Network Based, Speaker-Independent, Large Vocabulary, Continuous Speech Recognition System (ESPRIT Long Term Research Programme 1992-1995)