Advanced Audio Manipulation Algorithims Resource

[da peppester]

I am looking for a site when I can find links to or webpages on algorithms relating to advanced audio manipulation (if possible, some examples in c++).

What I mean by advanced audio manipulation is I have an audio file converted to an array of the frequency of the sound at given intervals of time (on the order of milliseconds/microseconds). Like so:

struct simpile_audio_sound {
    double interval;
    float* array;
}

And I am interested in a list of algorithms to manipulate that array of frequencies. Like, for an example of the bare minimal of what I hope to implement, separating out different people's voices (if that's what the current audio array contains), filtering out some of the voices based on if their tone is similar to the rest of the group, and recombine the voices of the people accordingly.

[RealGiganitris]

That's pretty damn advanced, why do you want to do this project? Can I suggest a project idea that I want to do but don't have the time for?

[htirwin]

The first part is actually pretty simple, probably the most basic and fundamental audio processing algorithm, FFT. There are some fairly easy to use open source projects. I used kiss fft before and it worked great for me.

https://sourceforge.net/projects/kissfft/

The next part is not easy. Can you be more clear what you mean by separate peoples voices? Detecting / determining who the speaker is, is not so hard. Removing just one out of many persons voices from an audio wave, I think would be very difficult in the case where multiple people are talking at once. In either case, I think you will use machine learning, with the features being the frequency distributions (part of the fft output) associated with peoples voices.

[da peppester]

RealGiganitris, I am very intrigued about your wonderful idea, I have sent you a private message about it so we can discus your stupendously wonderful idea. But, it is sorta off-topic from this thread, so lets not discuss any more of it here. If anyone else has any awesome ideas, I would love to hear about them, you can private message me, and I will love giving you full credit for your amazing idea.

Thank you so very much htirwin for that awesome link, you are so very incredibly helpful. Thank you so very much. Also, to be more specific about what I meant by people's voices, that was just a random off-the-top-of-my-head example to give you a good perspective on how advanced I am going with audio manipulation.