CNS Speech Lab
Speech Production

The DIVA Model
Model Demos
Neural Correlates

Speech Perception

Sound Categorization
Audio-Visual Speech

Brain Imaging

Overview



Department of
Cognitive & Neural Systems

Babbling

Demo 1: A portion of the babbling phase that allows DIVA to learn to produce vowels. (907 KB MPEG Movie)

Vowel Demonstrations

Demo 2: DIVA producing the vowels in "bet", "beet", "bat", "but", and "boot". (909 KB MPEG Movie)
Demo 3: DIVA producing the same vowels as in Demo 2, without using it's jaw. The model was never trained using a blocked jaw, which requires very different movements to produce the same sounds than as opposed to the unblocked case. (915 KB MPEG Movie)
Demo 4: DIVA producing the same vowels as in Demo 2, with 2/3 of tongue mobility removed. (884 KB MPEG Movie)
Demo 5: DIVA producing the same vowels as in Demo 2, without using it's jaw, lips, or larynx. The model produces sounds reasonably close to the targets, much like a ventriloquist. (887 KB MPEG Movie)

Consonant Demonstrations

Demo 6:DIVA producing the sequence abi (76 KB AVI)
Demo 7:DIVA producing the sequence api (76 KB AVI)
Demo 8:DIVA producing the sequence agi (76 KB AVI)


The DIVA Model
Hypothesized Neural Correlates of DIVA