论文信息 - On the Generalization Capability of Multi-Layered Networks in the Extraction of Speech Properties

On the Generalization Capability of Multi-Layered Networks in the Extraction of Speech Properties

The paper describes a speech coding system based on an ear model followed by a set of MultiLayer Networks (MLN). MLNs are trained to learn how to recognize articulatory features like the place and manner of articulation. Experiments are performed on 10 English vowels showing a recognition rate higher than 95% for new speakers. When features are used for recognition, comparable results are obtained for vowels and diphthongs not used for training and pronounced by new speakers. This suggests that MLNs suitably fed by the data computed by an ear model have good generalization capabilities over new speakers and new sounds.

[1] Geoffrey E. Hinton,et al. Learning sets of filters using back-propagation , 1987 .

[2] Lokendra Shastri,et al. Learning Phonetic Features Using Connectionist Networks , 1987, IJCAI.

[3] Yoshua Bengio,et al. Programmable execution of multi-layered networks for automatic speech recognition , 1989, CACM.

[4] S. Seneff. A joint synchrony/mean-rate model of auditory speech processing , 1990 .

[5] Victor W. Zue,et al. Some phonetic recognition experiments using artificial neural nets , 1988, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.

[6] Pietro Laface,et al. Parallel Algorithms for Syllable Recognition in Continuous Speech , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.

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

[8] Alex Waibel,et al. Phoneme recognition: neural networks vs. hidden Markov models vs. hidden Markov models , 1988, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.