论文信息 - FOR A CASCADE/PARAUEL FORMANT SPNTHESIZER

FOR A CASCADE/PARAUEL FORMANT SPNTHESIZER

Neural networks could help establish the generalizations inherent in text-to-speech synthesis rules. Such networks are showing the ability to learn the rules for resynthesizing a finite vocabulary and their degree of success in dealing with new situations depends on their training methodology and the size and composition of the training corpus. The cascade/parallel formant synthesizer model has been chosen as a case study.

Michael S. Scordilis | John N. Gowdy

[1] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .

[2] David B. Pisoni,et al. Text-to-speech: the mitalk system , 1987 .

[3] Akio Tanaka,et al. Speech Production Using A Neural Network with a Cooperative Learning Mechanism , 1988, NIPS.

[4] Bishnu S. Atal,et al. Speech synthesis by linear interpolation of spectral parameters between dyad boundaries , 1979 .

[5] John N. Gowdy,et al. Neural network based generation of fundamental frequency contours , 1989, International Conference on Acoustics, Speech, and Signal Processing,.

[6] Joseph P. Olive,et al. Speech resynthesis from phoneme-related parameters. , 1975 .

[7] H. Kucera,et al. Computational analysis of present-day American English , 1967 .