0
0

Hi! :)

I’d like to record what my microphone is hearing, and I’d like to get the spectrum of this (in order to recognize the notes).
Okay, I can recognize the notes and I can record the sound, but I can’t do this at the same time.

I know there’s 2 examples: pitchdetection and recordtodisk that are doing pretty much what I want (except for pitchdetection, there can be several notes at the same time), I’ve worked a lot with it.
But I can’t [i:1t06goba]merge[/i:1t06goba] the 2 example. I don’t know really why, maybe I’m lost with the FMOD_fuctions

Can somebody explain me quickly how to do this properly?
(Just get the spectrum while I’m recording, I know how to deal with the spectrum array)

Thanks!!!

  • You must to post comments
0
0

I think the piece of the puzzle you are missing here is playing the recorded voice. If you allow a short pause between System::recordStart and System::playSound on the recording sound you can playback what you recording with that slight delay for buffering. You can now use Channel::getSpectrum on the playing FMOD::Sound.

  • You must to post comments
0
0

I’d recommend taking a look at the pitch detection example, it combines recording with use of getSpectrum.

  • You must to post comments
0
0

Hi mathew,

I am trying the same thing as libljhj. For my project I try to get a sound with the microphone and getting the FFT with GetSpectrum. At the end it has to be a program a little like Shazam (a little like …)
I have seen your post and I don’t understand how to do it.
I use recordStart for 10 seconds after I make a little Sleep (about 5 ms) and I run playsound. But with GetSpectrum() I always have the same numbers in my array. So I don’t know if these are default values or what.

Thanks

  • You must to post comments
Showing 3 results
Your Answer

Please first to submit.