|
Professor Nelson Morgan is the former director of the International Computer Science Institute (ICSI), where he was also the Speech Group leader.〔(SLTC Newsletter ), IEEE Signal Processing Society, May 2012〕 He is also a professor in residence (emeritus) of electrical engineering and computer science at the University of California, Berkeley.〔(Author biography ), ''Speech and Audio Signal Processing'', Wiley Publishing, 2011〕 Morgan was born in 1949 and received his PhD as an NSF fellow from UC Berkeley in 1980.〔(NSF Graduate Research Fellows ). Retrieved 28 February 2012〕 He founded ICSI's Realization Group, which later become known as the Speech Group, in 1988. He served as director of ICSI from 1999 through 2011.〔(Interview with Dr. Roberto Pieraccini, Director of ICSI at Berkeley, USA ), KI - Künstliche Intelligenz, SpringerLink, published online June 29, 2012〕 He is the co-inventor of the Relative Spectral (RASTA) approach to speech signal processing, first described in a technical report published in 1991.〔(RASTA-PLP Speech Analysis ), Hynek Hermansky, Nelson Morgan, Aruna Bayya, and Phil Kohn, ICSI Technical Report TR-91-069, December 1991〕〔(RASTA-PLP Speech Analysis Technique ), Hynek Hermansky, Nelson Morgan, Aruna Bayya, and Phil Kohn, Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP-92), Vol.1, No., pp. 121-124 March 1992. doi: 10.1109/ICASSP.1992.225957〕 In 1993, Morgan and Herve Bourlard published their work on the hybrid system approach to speech recognition, which uses neural networks probabilistically with Hidden Markov Models (HMMS).〔''(Connectionst Speech Recognition: A Hybrid Approach )'', Nelson Morgan and Herve Bourlard, Springer International Series in Engineering and Computer Science, Vol. 247, 1993〕 The system improved automatic speech recognition techniques based on HMMs by providing discriminative training, incorporating multiple input sources, and using a flexible architecture able to accommodate contextual inputs and feedbacks. The work has been described as "seminal.".〔(Hybrid HMM/Neural Network Based Speech Recognition in Loquendo ASR ), Roberto Gemello, Franco Mana, and Dario Albesano〕 Morgan won the 1996 IEEE Signal Processing Magazine Best Paper Award for a paper with Bourlard.〔(IEEE Signal Processing Magazine Best Paper Award Recipients ), IEEE Signal Processing Society〕 Morgan was the principal investigator of the IARPA-funded project Outing Unfortunate Characteristics of HMMs, which sought to identify problems in automatic speech recognition technology.〔(Speech Tech Blog ), Michele Masterson, June 20, 2012〕 He has more than 200 publications, including three books.,〔''Speech and Audio Signal Processing'', Ben Gold and Nelson Morgan, Wiley Publishing, 1999〕〔''(Speech and Audio Signal Processing, Second Edition )'', Ben Gold, Nelson Morgan, and Dan Ellis, 2011〕 He is a fellow of the IEEE〔(IEEE Fellows ). Retrieved 28 February 2012.〕 and the International Speech Communication Association.〔() ISCA 2010 Fellows. Retrieved 28 February 2012〕 He is on the editorial board of ''Speech Communication Magazine'', of which he is a former co-editor-in-chief.〔(Speech Communication Editorial Board ), Elsevier Publishing.〕 == References == 抄文引用元・出典: フリー百科事典『 ウィキペディア（Wikipedia）』 ■ウィキペディアで「Nelson Morgan」の詳細全文を読む スポンサード リンク
|