Pika Dynamic Captions

Burn word-timed captions into a talking-head clip: hand over the video, the words with their timings, a caption style and a colour, and get the same clip back with the captions rendered over the picture. The output keeps the source's audio and length, and its size up to 2160 pixels on the long side (larger clips are scaled down). SDR sources up to 1,200 seconds; convert PQ/HLG HDR to SDR first. Billed per output second.