I’ve finally done it! I’ve connected the pieces that I’ve been working on for the last 2 months and I can now generate audio out of data in our defect database at work.
Yep, that’s right. I really am crazy enough to try and implement something like that. …and I did.
The current algoriths used are still very basic. From generating basic sinusoids of various frequencies to composing a musical score the code that I’ve written just barely gets the job done. This leaves lots of room for improvement.
For example, some of the algorithms still leave pops and clicks in the resulting waveform and I can’t really handle notes that occur at the same time very cleanly right now. It’s very computationally intensive and the algorithms used here really don’t help. But the point is that it works.
Everything else can be improved bit by bit down the road my focus was to implement the basic requirements so that I could "hear" what our defect database could sound like.
I will try and post some example tracks at some point down the road.