Continuous multi-band speech recognition using Bayesian networks
Tóm tắt
Using the Bayesian networks framework, we present a new multi-band approach for continuous speech recognition. This new approach has the advantage of overcoming all the limitations of the standard multi-band techniques. Moreover, it leads to a higher fidelity speech modeling than HMMs. We provide a preliminary evaluation of the performance of our new approach on a connected digits recognition task.
Từ khóa
#Speech recognition #Bayesian methods #Hidden Markov models #Random variables #Automatic speech recognition #Decoding #VocabularyTài liệu tham khảo
daoudi, 0, A new approach for multi-band speech recognition based on probabilistic graphical models, ICSLP’ 2000
moore, 1989, An introduction to the psychology of hearing
10.1109/ICASSP.2000.862013
zweig, 1998, Speech Recognition With Dynamic Bayesian Networks
jensen, 1990, Bayesian updating in recursive graphical models by local computations, Computational Statistics and Data Analysis, 269
10.1007/BF01890546
10.1109/ICSLP.1996.607154
bourlard, 0, A new ASR approach based on independent processing and recombination of partial frequency bands, ICSLP'96