preprocess.py

import librosa
import os
from sklearn.model_selection import train_test_split
from keras.utils import to_categorical
import numpy as np
from tqdm import tqdm
from SPECtogram import gimmeDaSPECtogram

DATA_PATH = "./data/"


# Input: Folder Path
# Output: Tuple (Label, Indices of the labels, one-hot encoded labels)
def get_labels(path=DATA_PATH):
    #labels = os.listdir(path)
    labels = ['up', 'down', 'left', 'right', 'one', 'two', 'three', 'four', 'stop', 'go']
    #print(labels)
    label_indices = np.arange(0, len(labels))
    return labels, label_indices, to_categorical(label_indices)



# Handy function to convert wav2mfcc
def wav2mfcc(file_path, max_len=11):
    wave, sr = librosa.load(file_path, mono=True, sr=None)
    wave = wave[::3]
    mfcc = librosa.feature.mfcc(wave, sr=16000)

    # If maximum length exceeds mfcc lengths then pad the remaining ones
    if (max_len > mfcc.shape[1]):
        pad_width = max_len - mfcc.shape[1]
        mfcc = np.pad(mfcc, pad_width=((0, 0), (0, pad_width)), mode='constant')

    # Else cutoff the remaining parts
    else:
        mfcc = mfcc[:, :max_len]

    return mfcc



def save_data_to_array(path=DATA_PATH, max_len=12):
    labels, _, _ = get_labels(path)

    for label in labels:
        # Init mfcc vectors
        mfcc_vectors = []

        wavfiles = [path + label + '/' + wavfile for wavfile in os.listdir(path + '/' + label)]
        #i = 0
        for wavfile in tqdm(wavfiles, "Saving vectors of label - '{}'".format(label)):
            #if i == 3700:
            #    break
            mfcc = gimmeDaSPECtogram(wavfile)
            #print(mfcc.shape)
            mfcc_vectors.append(mfcc)
            #i += 1
        np.save(label + '.npy', mfcc_vectors)

# Run this to prepare data for tensorflow feeding


def get_train_test(split_ratio=0.92, random_state=42):
    # Get available labels
    labels, indices, _ = get_labels(DATA_PATH)

    print(labels[1:])

    # Getting first arrays

    X = np.load(labels[0] + '.npy')
    print(X)
    y = np.zeros(X.shape[0])

    # Append all of the dataset into one single array, same goes for y
    for i, label in enumerate(labels[1:]):
        x = np.load(label + '.npy')
        X = np.vstack((X, x))
        y = np.append(y, np.full(x.shape[0], fill_value=(i + 1)))

    print(X.shape[0])
    print(len(y))

    assert X.shape[0] == len(y)

    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=(1 - split_ratio), random_state=random_state, shuffle=True)

    return X_train, X_test, y_train, y_test


def prepare_dataset(path=DATA_PATH):
    labels, _, _ = get_labels(path)
    data = {}
    for label in labels:
        data[label] = {}
        data[label]['path'] = [path + label + '/' + wavfile for wavfile in os.listdir(path + '/' + label)]

        vectors = []

        for wavfile in data[label]['path']:
            wave, sr = librosa.load(wavfile, mono=True, sr=None)
            # Downsampling
            wave = wave[::3]
            mfcc = librosa.feature.mfcc(wave, sr=16000)
            vectors.append(mfcc)

        data[label]['mfcc'] = vectors

    return data


def load_dataset(path=DATA_PATH):
    data = prepare_dataset(path)

    dataset = []

    for key in data:
        for mfcc in data[key]['mfcc']:
            dataset.append((key, mfcc))

    return dataset[:100]

# print(prepare_dataset(DATA_PATH))