That's a hard thing to do because decoding the current frame can require data left over from decoding the previous two. libmpg123 manages even sample-accurate seeking very well - maybe you could try that.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
That's a hard thing to do because decoding the current frame can require data left over from decoding the previous two. libmpg123 manages even sample-accurate seeking very well - maybe you could try that.