audio chroma features detection, usage: "python3.6 chroma.py -d data/"

chroma.py

from __future__ import print_function
import sys
import os
import os.path
from os.path import basename
import argparse
import json
import hashlib
import librosa
import librosa.display
import matplotlib.pyplot as plt

from model.track import Track
import codecs, json

# args
parser = argparse.ArgumentParser("processor")
parser.add_argument("-f", help="file to process", type=str, dest="filename")
parser.add_argument("-d", help="directory to process", type=str, dest="directory")

args = parser.parse_args()
filename = args.filename
directory = args.directory

files = []
if directory == None:
  files.append(filename)

if filename == None:
  for f in os.listdir(directory):
    files.append(directory + os.fsdecode(f))

for filename in files:
  t1 = Track()
  t1.file_hash = hashlib.md5(filename.encode('utf-8')).hexdigest()
  t1.file_path = filename
  t1.file_name = basename(filename)

  target = "out/" + os.path.splitext(t1.file_name)[0] + ".png"

  if os.path.exists(target):
    print("skipping: " + t1.file_name)
    continue

  print("processing: " + t1.file_name)
  open(target, 'a').close()

  # process
  y, sr = librosa.load(filename)
  t1.r_length = librosa.get_duration(y=y, sr=sr)
  onset_env = librosa.onset.onset_strength(y, sr=sr)
  y_harm, y_perc = librosa.effects.hpss(y)
  chroma = librosa.feature.chroma_cqt(y=y_harm, sr=sr)

  # plot and save to file
  plt.figure(figsize=(100,8))
  librosa.display.specshow(chroma, sr=sr, x_axis='time', y_axis='chroma', vmin=0, vmax=1)
  plt.title('Chromagram')
  plt.colorbar()
  plt.tight_layout()
  plt.savefig(target)