I think that in some editing tasks it would be of great help to have intelligible audio in fast forward modes.
Currently, in fast forward mode you get the concatenation of audio fragments played at regular speed with pieces missing in between them which completely breaks the ability to understand what the audio is about and makes the audio in fast forward modes rather useless.
By comparison, some applications from different domains have shown how useful it is to play something at increased speed maintaining the intelligibility of the audio, particularly when it is a speech. A notable example is whatsapp where many users routinely take advantage of the ability to play audio messages at 1.5x or 2x to save time and to get to the part they are interested in rapidly. When correctly processed the audio track at 2x remains fine and it is perfectly possible to understand what is being said, which music is being played, etc.
In Shotcut, having a fast forward mode with intelligible audio would make some editing tasks very much faster, since it would let one rapidly go through a video and identify the parts that need editing. A typical case is when editing talks or lecture videos.