support pinning HF model revision via DOTS_TTS_MODEL_REVISION env var

Revert "use reflect-pad extra patch on prompt audio instead of regenerate-and-drop"

bump default num_steps to 16

use reflect-pad extra patch on prompt audio instead of regenerate-and-drop

switch default execution mode to generate

add voice cloning tips card to UI

add constraint on both prompt text and text to generate

update constraints for better performance

add some constraints for better performance

set default generation time to 60s

copy from /tmp to persist bucket to avoid gradio error

refine Gradio placeholder translation

change generated dir to huggingface space mounted bucket

add an import

add inference code with AOTI support for hf space

initial commit