I am currently using Whisper AutoSRT for processing large volumes of audio and video files for subtitle generation and translation. Given the computational demands of this task, especially with high-resolution or lengthy media files, the ability to utilize multiple GPUs concurrently would significantly enhance performance and efficiency.