Using the FFT library, is it possible to get the beats per minute? If so what would I type exactly?
Hello,
Please include references when discussing a library.
FFT is part of the Processing sound library:
I have a deep voice and was beeping like a metronome to test some code out.
I actually made “beep” sounds.
Steps:
- Started with:
FFT / Libraries / Processing.org - focused only on low end of the spectrum from the array
println() / Reference / Processing.org used to see data to get a sense of it - grabbed the time when there was a “beep” (peak power at the frequency range of interest)
millis() / Reference / Processing.org - calculate the difference between times
- flags (true\false) were used in code to keep track of peaks and lows so I was able to calculate time differences
I was able to get the time between “beeps” in this first attempt.
It would require a lot more work to clean this up and work with a broader spectrum of sound with a “beat” hidden in there.
Some of the filters look like something to try; depends on what the “beat” is that you are filtering.
Sometimes you have to write your own code from what is available in existing libraries.
This is an academic exercise from what I have gleaned from your other posts so I will not be sharing code.
:)
What sensor are you using?
Hi man, sure I will keep that in mind next time.
I will check out the links.
[quote=“glv, post:2, topic:22965”]
It would require a lot more work to clean this up and work with a broader spectrum of sound with a “beat” hidden in there.
Sounds like it
At that point why to use Processing, I might as well use something like C as this will give me more low level functionality. I am forced to use Processing.
Not sensors like I want to take the audio file as the input.
I understand how this works, but how would I implement it with the FFT library?