Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

There are so many concert recordings of awesome performances that sound like crap because they are audience tapes.

Often before a performance recorded music is played and captured in the audience recordings.

Would it be possible to train a model on an archive of these concert audience recordings of studio recordings paired with the original studio recordings to develop a system to “clean up” audience recordings?



Maybe... however, most bands tend to drift a LOT from the studio versions of songs while touring.


I'm referring to the pre-concert, when they're just playing a mixtape of whatever through the board. Compare studio recording of whatever and the audience recording.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: