ProMode learns a prosody embedding from partially masked audio and text, improving F0 and energy prediction over baseline style encoders.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ProMode: A Speech Prosody Model Conditioned on Acoustic and Textual Inputs
ProMode learns a prosody embedding from partially masked audio and text, improving F0 and energy prediction over baseline style encoders.