How do I do mathematical analysis on microphone audio information from pyaudio?

Question

I am trying to obtain the pitch, the level, and other stuff from my microphone using Python. I am trying using pyaudio, but I read that I need to use other libraries to generate the maths. I can’t find how mix pyaudio with the math library.

Answer 1

I have performed basic signal processing (FFT etc) using numpy/scipy directly on the audio buffer without too much trouble. Be careful, you need to tell numpy about the buffer type: float or int, long or short, signed/unsigned ...

Answer 2

You can use the standard Python module audioop to determine the power/loudness of an audio sample, with the audioop.rms function.