Compute Features¶
Shrimply only shows features and models advertised by the selected compute server. Connect a server in , then choose what you want to do:
Transcription turns selected audio into timed caption clips.
Text to Speech creates speech from text or existing captions.
Video Segmentation follows a selected subject through a video.
Voice Conversion changes recorded speech with an installed voice model.
3D Camera Tracking recovers a 3D camera path from a visual track.
Video Generation creates video from text, images, or references.
Models normally download the first time they are used. Keep the compute server running until the current operation finishes.