Reversing a Video: What Happens to the Audio and the File Size
Reversing a video sounds like it should be as cheap as reversing a trim, just play the same data backward. It isn't, and the reasons explain two things people are consistently surprised by: the audio usually disappears, and the resulting file can end up larger than the original.
Why reversing forces a full re-encode
Trimming a video can sometimes just copy a byte range, cutting between two points doesn't require touching the pixels. Reversing can't take that shortcut, because modern video codecs don't store every frame as a complete image. Most frames are stored as a difference from the frames around them, which only makes sense when decoded in forward order, a frame defined as "what changed since the last frame" is meaningless in isolation. To reverse a video, every frame first has to be fully decoded into raw pixel data, the whole sequence then gets reordered back to front, and the result has to be re-encoded from scratch as a brand new compressed stream. There's no way to reverse the compressed bytes directly. That full decode-reorder-encode pipeline is why reversing a video is noticeably more computationally expensive than a simple trim or cut.
Why the audio usually doesn't survive
Audio has the same directional problem as video, reversed speech or music doesn't play back as "the same audio, backward and still listenable," it plays as a distorted, backward mess, speech in particular becomes unintelligible noise almost immediately. Reversing audio in sync with reversed video also means re-timing it against a completely re-encoded video stream, real added complexity for an output that, once reversed, usually isn't something anyone actually wants to hear anyway. That's why reverse tools commonly drop the audio track entirely rather than ship a reversed track nobody would use. Kit-Bin's Reverse Video tool follows the same reasoning: it decodes and re-encodes the video frames only, with no audio track carried through to the output. If you specifically want the audio from a clip,Mute Video or an audio extraction pass on the original (unreversed) file is the way to get it, reversed video isn't the place to expect usable sound.
Why the file can come out bigger
Because a reversed output is always a fresh encode rather than a copy of the original compressed stream, its file size depends entirely on the encoding settings used for that new pass, not on the original file's size. If your source clip happened to be encoded efficiently, at a low bitrate, with a modern codec tuned well for its content, a re-encode at a different quality target can easily land larger, even though visually nothing looks different. This isn't a bug or a sign something went wrong, it's inherent to the fact that reversing requires a full re-encode, there is no version of "reverse the video" that can also guarantee "and keep the exact original file size," because the exact original compressed bytes cease to exist the moment the video gets decoded to reverse it. If the resulting file is larger than you'd like, Compress Video afterward brings it back down.