Sign in to follow this  

Audio wavetable synthesis

This topic is 2586 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I'm trying to understand how wavetable synthesis work.
Let's say I have a wavetable with 5000 samples of a piano, and that samples represent a note, let's say, middle C.

Now, I want to synthetize another note, for example, an A above middle C, and write the result to a sound buffer.
If I understand correctly, I need to resample the original samples at a different rate.

Frequency of middle C: 261.63
Frequency of A above it: 440
440 / 261.63 = 1.68176432

This means the A needs to be resampled 1.68176432 times faster than C.
So I would for example, do something like

float sourcePos = 0;
for (int i=0; i<5000; i++)
{
dest[i]=samples[(int)sourcePos];
sourcePos += 1.68176432;
if ((int)sourcePos>5000) sourcePos -= 5000; // wrap around
}

Is my logic right until now?
(Of course, if this correct, then the algoritm should be improved with some kind of interpolation)

Also I didn't find a audio programming subforum here, so I hope this is the right place for this topic.

Share this post


Link to post
Share on other sites
Well, I just implemented the algoritm I described and it sounds horribly, even just one semitone up or down...

This code works fine when Dx = 1.0, every other value outputs strange sound data, for example, I as using an ocarina sample of a A6 note, and when trying an A#6 (one semitone up), instead of sounding like an ocarina, sounds like something between a motor and a very big bell.


float Dx = Frequencies[Note] / Frequencies[_BaseNote];
float Pos = 0.0;
while (samples>0)
{
uint16 N = _sample[(int)(Pos)];
*Dest = N; Dest++; // left channel
*Dest = N; Dest++; // right channel
Pos += Dx;
if (Pos>_LoopEnd)
Pos -= (_LoopEnd-_LoopStart);
Samples--;
}


(Ignore any sintax error in the code, I just converted it into C++ now, I'm using a different language)

EDIT: Fixed, it was actually a bug in my implementation, I was not sampling correctly when the position wrapped around.

[Edited by - Relfos on November 12, 2010 6:06:31 PM]

Share this post


Link to post
Share on other sites

This topic is 2586 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this