Takes a voice embedding, and target language, and localizes the voice to the specific language. For certain languages, this improves the quality of the accent.
Pricing: This endpoint costs 225 characters of utilization per call.
The version of the Cartesia API to use.
Target language, which the output embedding will be localized to. Options: en
(English), de
(German), es
(Spanish), fr
(French), ja
(Japanese), pt
(Portuguese), zh
(Chinese).
Speaker gender of the original speaker, used to accurately localize voice.
Target accent/dialect, which the output embedding will be localized to. Currently only supported for language en
. Options: au
(Australian), in
(Indian), so
(Southern [American]), uk
(British), us
(American).