Freesound Forums

Hey guys, I'm in the middle of creating an app that automatically tags sound clips and lets people generate sounds by typing in a sentence. The problem is, I'm knee deep in coding and I have no idea if anyone would be interested in it! So I would very much appreciate your thoughts.

Would you use it?

Is there something better that I could do that involves machine learning and sound?

Premise:
The app would do two things 1. automatically tag soundclips with appropriate tags and 2. generates sound clips. In order to generate sound clips the user would enter a string of tags and a machine learning algorithm would try to generate a sound that best matches those clips. So for instance, you could type in "birds, stream" and it should generate bird calls and the sound of a stream even if there is no example of such a sound clip.

I was also thinking that I could probably speed up the approval process by training a machine learning algorithm to flag bad quality sounds. I'm not really in contact with the owners of the site though.

Any feedback would be much appreciated! Thanks in advance!