ISCA Archive ICSLP 1992
ISCA Archive ICSLP 1992

A hidden Markov model structure for the acquisition of speech by machine, ASM

Frank Fallside

A structure for the acquisition of speech by machines, asm, employing neural networks, for the 'simultaneous' training of recognition and synthesis has been presented [1]. In the present paper it is shown that there is an equivalent structure employing HMMs. By analogy with recent results relating HMM and neural network architectures it is seen that the two methods for asm are equivalent.