Artificial evolution for computer graphics
Proceedings of the 18th annual conference on Computer graphics and interactive techniques
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Voice Conversion Using Interactive Evolution Of Prosodic Control
GECCO '02 Proceedings of the Genetic and Evolutionary Computation Conference
Advances in evolutionary computing: theory and applications
Advances in evolutionary computing: theory and applications
Evolutionary computation: comments on the history and current state
IEEE Transactions on Evolutionary Computation
Hi-index | 0.00 |
This paper reports the results of evaluation experiments performed in relation to text dependency in voice quality conversion using interactive evolution. It is important to investigate beforehand whether text dependency exists when considering engineering applications of voice quality conversion technology. In these experiments, for both natural speech recorded with a microphone and synthetic speech generated from text data, prosodic conversion coefficients previously determined for each conversion target by the interactive evolution technique was applied to different text data for the same speaker, and subjects having no knowledge of the purpose of the experiments were asked to evaluate the speech after conversion. We confirmed that prosodic conversion coefficients determined by the interactive evolution technique, while exhibiting speaker dependency, is not text dependent.