ajr@eng.cam.ac.uk (Tony Robinson) (10/09/90)
I'm after pointers to the latest work on connectionist large-vocabulary speaker-independent continuous speech recognition. If you drop one of the constraints, I can find: constraint dropped reference connectionist Kai-Fu Lee, "Automatic Speech Recognition: The Development of the SPHINX System", Kluwer 1989. speaker-independent N Morgan and H Bourlard, "Continuous Speech Recognition Using Multilayer Perceptrons with Hidden Markov Models", NIPS-89 and ICASSP-90 large-vocabulary Michael Franzini, Kai-Fu Lee, and Alex Waibel, "Connectionist Viterbi Training: A New Hybrid Method for Continuous Speech Recognition", ICASSP-90. continuous speech Masanori Miyatake, Hidefumi Sawai, Yasuhiro Minami and Kiyohiro Shikano, "Intgrated Training for Spotting Japanese Phonemes Using Large Phonemic Time-Delay Neural Networks", ICASSP-90. [ICASSP-90 is the proceedings of the 1990 International Conference on Acoustics, Speech and Signal Processing (IEEE Signal Processing Society) and contains many other relevant papers.] Can anyone point me to any work that satisfies all the constraints, or any better references than those above (especially if they were performed on the DARPA TIMIT or Resource Management databases), or to connectionist speaker-independent recognition of a full phoneme set. A summary will appear given enough response to summarise. Thanks! Tony Robinson.