one thing:
00:53:687 (53687|0,54318|0) | these two notes were the ones that made me think that these were not representing vocals, the 1st is very detached from any vocal which if it was supposed to then i think it should be at 00:53:529 instead since thats where one actually starts, the 2nd is also in part the same as vocal at 00:54:160 ends too short for it to be represented so think this should just be a jump instead
01:31:872 (91872|3,92059|2) | related, can you move this to the left by one column for the same reason