That makes sense. It’s definitely a long shot technique, as the music is spread across 5 channels of film and the soundtrack presumably is merely stereo with two different waveforms. You’d have to hope that the stereo version is exactly the one used in the film and identify if the tracks have been separated between left and right speakers or merged into several more. This process would probably only work if you had two identical recordings of the same channel number and processed in the same way with the only difference being the addition of vocals in one of them.