An interactive genetic algorithm tunes arousal-valence coordinates per listener in emotional TTS, and personalized or culture-specific coordinates beat a generic U.S.-average baseline in small A/B tests.
Dataset and Training Setup Dataset:We constructed a 9-hour American English fe- male emotional speech dataset (22.05 kHz) by combining EX- PRESSO [27], EmoV-DB [28], and ESD [9]
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2026 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Beyond One-Size-Fits-All: Personalized and Culturally Adaptive Emotional TTS via Interactive Optimization of Individual Emotion Perception Spaces
An interactive genetic algorithm tunes arousal-valence coordinates per listener in emotional TTS, and personalized or culture-specific coordinates beat a generic U.S.-average baseline in small A/B tests.