Supervised and Unsupervised Feature Extraction from a Cochlear Model for Speech Recognition
BROWN UNIV PROVIDENCE RI INST FOR BRAIN AND NEURAL SYSTEMS
Pagination or Media Count:
We explore the application of a novel classification method that combines supervised and unsupervised training, and compare its performance to various more classical methods. We first construct a detailed high dimensional representation of the speech signal using Lyons cochlear model and then optimally reproduce its dimensionality. The resulting low dimensional projection retains the information needed for robust speech recognition.
- Anatomy and Physiology
- Test Facilities, Equipment and Methods
- Voice Communications