Philip Jackson

Philip Jackson holds an MA in Engineering from Cambridge University and a PhD in Electronic Engineering from the University of Southampton, UK. He worked in industry for Ultra Electronics on the world’s first active noise controlled cabin for commercial aircraft [JacksonRoss:ASME:96], leading to the Queen’s Award for Technology in 1996. After a successful postdoc at the University of Birmingham, he joined the University of Surrey as lecturer in 2002, and now heads the machine audition group within CVSSP, which has two academics and a dozen research staff and students.

Dr Jackson has relevant expertise along three themes: speech production, articulatory modeling, and the combination of audio with other modalities. Recent research projects have involved multimodal analysis, synthesis and recognition, using pseudo-articulatory and audio [JacksonEtAl:El-Lett:02, RussellJackson:CSL:05], articulatory and audio
[JacksonSingampalli:Specom:09], 2D and 3D visual motion capture data and audio [TurkmaniEtAl:IS:07, EdgeEtAl:JASMP:09], and audio-visual fusion for expressed emotion recognition [HaqJackson:AVSP:08]. His speech production research has contributed to use of harmonic structure for noise-robust ASR [JacksonEtAl:ES:03], and characterisation of voiced fricative speech [JacksonShadle:JASA:00, JacksonShadle:T-SAP:01]. With regard to design of subjective evaluation protocols, Dr Jackson also has experience of spatial sound quality [ConettaEtAl:AES:08], detection, discrimination and cue trading of fine temporal structure [PincasJackson:ASA:08a, PincasJackson:ASA:08b], and use of different modalities in human affect classification [HaqJackson:AVSP:08]. His project output (e.g., from BALTHASAR, DANSA, Dynamic Faces and QESTRAL projects) shows a record of high quality research, including the possibility of exploitation following several US and UK patent filings with industrial support from the BBC and Bang & Olufsen.

His current interests include production and perception of voiced fricatives, models of speech articulation and multimodal recognition of expressed emotions. He has published over 40 scientific papers in high-quality academic journals and conference proceedings, and reviews for journals and conferences such as Journal of the Acoustical Society of America, IEEE Signal Processing Letters, InterSpeech and ICASSP, and is on the editorial board of Computer Speech and Language.