frequency analysis

Status
Not open for further replies.
Hello There,

I would like to implement one microphone to do frequency analysis between 20Hz – 20kHz.



Is teensy 3.2 capable of sampling this frequency rate?



may you know some microphone compatible with teensy 3.2 with this requirements?




i'll appreciate all the support for my project
 
The Nyquist theorem requires that the sampling frequency is at least the double of the maximum audio frequency to be sampled. As the documentation of the audio library states, it works with a fixed 44.1kHz sampling rate, so your desired frequency range is covered by default. Read the doc and the tutorials, use the GUI to run a few examples and you'll see.
 
I recommend you start with this tutorial:

https://www.pjrc.com/store/audio_tutorial_kit.html

It has a 31 page PDF and we made a 45 minute walkthrough video which shows all the steps. The FFT material is covered in part 3-2, starting on page 24.

To specifically answer your question:

i mean how much cycles have to make the device for every acquisition?

With a 1024 point FFT, 1024 audio samples are collected. However, usually a window shape is applied to the data to prevent spectral leakage, which is described in the tutorial. The window shape means the samples in the center of those 1024 are weighted much more than those on the ends. The actual acquisition for each FFT is the center portion of each 1024 samples (at 44.1 kHz).

The audio library also has an excellent note frequency analysis written by Collin Duffy, using the YIN algorithm. Like everything in the library, it's documented in the design tool.

https://www.pjrc.com/teensy/gui/?info=AudioAnalyzeNoteFrequency

This feature isn't covered in the tutorial, so I would recommend you first do the tutorial and try FFT. But for some types of applications (like detecting the actual fundamental frequency of musical instruments), YIN works much better than FFT. Hopefully after you have used the tutorial, you will gain an understanding of how to use the design tool and use the other objects which aren't explicitly covered in the tutorial PDF.
 
my project is about, acquire the sound and send it to a computer via WiFi.

the objective is make frequency analysis, but i don't know what's the better way to do that.

I acquire sound, make frequency analysis and send the data to the computer o i acquire sound, send this sound to computer and make frequency analysis in the computer?
What do you think is better?
 
@Spartan_001, it would be helpful for us to know more about the context of your question. Some questions to help us understand

1. What is your experience level with projects like this? Are you a student, hobbyist, etc? Do you have previous experience any similar projects?

2. Why do you want to make this project? School assignment? Just for fun and learning? For a job?

3. What is the end goal? What would you use this circuit for?
 
Status
Not open for further replies.
Back
Top