PCA of repeated synthesized utterances with fixed inputs can reveal controllable prosodic features that can be enrolled as new prompts via fine-tuning.
Given text, TTS models pre- dict a probable speech rendition under the statistics of the train- ing data
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
RepeaTTS: Towards Feature Discovery through Repeated Fine-Tuning
PCA of repeated synthesized utterances with fixed inputs can reveal controllable prosodic features that can be enrolled as new prompts via fine-tuning.