See it in Search
This page is a preview of the following resource. Continue onto eagle-i search using the button on the right to see the full record.
Cross-fading weight function algorithm
eagle-i ID
http://ohsu.eagle-i.net/i/0000013b-1fb8-d8bd-c825-3bd680000000
Resource Type
Properties
-
-
Resource Description
-
"We propose a new algorithm that uses a unit-dependent trainable parameterized cross-fading weight function to generate more natural-looking formant trajectories and, it is hoped, better-sounding output speech. The proposed algorithm:
- uses a perceptually-based objective function to capture
differences between cross-faded and natural trajectories
across the whole region of the phoneme, and
- uses phoneme identity, prosodic contexts, and acoustic
features of the units to predict optimal cross-fading parameters to generate more natural formant trajectories."
-
-
Used by
-
Center for Spoken Language Understanding
-
-
Related Publication or Documentation
-
Perceptual Cost Function for Cross-fading Based Concatenation
-
-
Developed by
-
Kain, Alexander, Ph.D.
-
-
Developed by
-
van Santen, Jan P.H., Ph.D.
-
-
Developed by
-
Miao, Qi