Create Audio Spectrograms with Python

Translation: de
Sun, 28 Jul 2013
Warning! The information on this page is heavily outdated. For a better way to visualize log-frequency spectrograms in Python, I recommend the excellent notebooks on Fundamentals of Music Processing, in particular the notebook on log-frequency spectrograms.

Here it's about creating spectrograms from WAVE files with Python, including decibel converted values and logarithmic scaled frequency axis. It uses NumPy, SciPy and matplotlib. Since this is a old code snippet, it requires a rather old version of NumPy, e.g. 1.10.4. If you are using Anaconda, you can create a working environment with conda create -n spec python=3 numpy==1.10.4 scipy matplotlib. If you have comments, improvements or find it useful, please conctact me.

An example: the spectrogram of the first 10 seconds of the song »Die Schnitzelbank« (sung by the Manhattan Quartett, likely in the 1920s, source: »American Memory« from the Libary of Congress, Washington, D.C.).

#coding: utf-8
""" This work is licensed under a Creative Commons Attribution 3.0 Unported License.
    Frank Zalkow, 2012-2013 """

import numpy as np
from matplotlib import pyplot as plt
import as wav
from numpy.lib import stride_tricks

""" short time fourier transform of audio signal """
def stft(sig, frameSize, overlapFac=0.5, window=np.hanning):
    win = window(frameSize)
    hopSize = int(frameSize - np.floor(overlapFac * frameSize))

    # zeros at beginning (thus center of 1st window should be for sample nr. 0)
    samples = np.append(np.zeros(np.floor(frameSize/2.0)), sig)    
    # cols for windowing
    cols = np.ceil( (len(samples) - frameSize) / float(hopSize)) + 1
    # zeros at end (thus samples can be fully covered by frames)
    samples = np.append(samples, np.zeros(frameSize))

    frames = stride_tricks.as_strided(samples, shape=(cols, frameSize),
    frames *= win

    return np.fft.rfft(frames)    

""" scale frequency axis logarithmically """    
def logscale_spec(spec, sr=44100, factor=20.):
    timebins, freqbins = np.shape(spec)

    scale = np.linspace(0, 1, freqbins) ** factor
    scale *= (freqbins-1)/max(scale)
    scale = np.unique(np.round(scale))

    # create spectrogram with new freq bins
    newspec = np.complex128(np.zeros([timebins, len(scale)]))
    for i in range(0, len(scale)):
        if i == len(scale)-1:
            newspec[:,i] = np.sum(spec[:,scale[i]:], axis=1)
            newspec[:,i] = np.sum(spec[:,scale[i]:scale[i+1]], axis=1)

    # list center freq of bins
    allfreqs = np.abs(np.fft.fftfreq(freqbins*2, 1./sr)[:freqbins+1])
    freqs = []
    for i in range(0, len(scale)):
        if i == len(scale)-1:
            freqs += [np.mean(allfreqs[scale[i]:])]
            freqs += [np.mean(allfreqs[scale[i]:scale[i+1]])]

    return newspec, freqs

""" plot spectrogram"""
def plotstft(audiopath, binsize=2**10, plotpath=None, colormap="jet"):
    samplerate, samples =
    s = stft(samples, binsize)

    sshow, freq = logscale_spec(s, factor=1.0, sr=samplerate)
    ims = 20.*np.log10(np.abs(sshow)/10e-6) # amplitude to decibel

    timebins, freqbins = np.shape(ims)

    plt.figure(figsize=(15, 7.5))
    plt.imshow(np.transpose(ims), origin="lower", aspect="auto", cmap=colormap,

    plt.xlabel("time (s)")
    plt.ylabel("frequency (hz)")
    plt.xlim([0, timebins-1])
    plt.ylim([0, freqbins])

    xlocs = np.float32(np.linspace(0, timebins-1, 5))
    plt.xticks(xlocs, ["%.02f" % l for l in
    ylocs = np.int16(np.round(np.linspace(0, freqbins-1, 10)))
    plt.yticks(ylocs, ["%.02f" % freq[i] for i in ylocs])

    if plotpath:
        plt.savefig(plotpath, bbox_inches="tight")