Welcome to the forum and the field!
This is a great first topic to begin with as well. I would recommend by reading this paper first which does a nice job of explaining both. In brief, amplicon sequence variants (ASVs) are, in their simplest form, a higher resolution version of OTUs, or 100% OTUs. That is to say the difference of a single nt will lead to designation of new ASV whereas with OTUs, reads are typically clustered at some % identity (most commonly 97%). Note that you can create ASVs then cluster them down to get OTUs, but you cannot go the reverse because once you have clusters there is no way of knowing what the original composition of those reads was. If you have raw data (.fastq) from which those OTUs were created from you can certainly re-analyse and start with ASVs.