{"id":1552,"date":"2021-04-06T13:01:35","date_gmt":"2021-04-06T07:31:35","guid":{"rendered":"https:\/\/dphi.tech\/blog\/?p=1552"},"modified":"2022-11-15T14:57:54","modified_gmt":"2022-11-15T09:27:54","slug":"audio-data-analysis-using-deep-learning-with-python-part-2","status":"publish","type":"post","link":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/","title":{"rendered":"Audio Data Analysis Using Deep Learning with Python (Part 2)"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">If you are an absolute beginner to audio data analysis, you should read <a href=\"https:\/\/www.theaidream.com\/post\/audio-data-analysis-using-deep-learning-with-python-part-1\" target=\"_blank\" rel=\"noreferrer noopener\"><u>Part 1<\/u><\/a> of this series first. Once you are comfortable with the concepts explained in that article, you can come back and continue with this article.<\/p>\n\n\n\n<figure class=\"wp-block-gallery has-nested-images columns-default is-cropped wp-block-gallery-1 is-layout-flex wp-block-gallery-is-layout-flex\">\n<figure class=\"wp-block-image size-large\"><img fetchpriority=\"high\" decoding=\"async\" width=\"570\" height=\"289\" data-id=\"1553\" src=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/unnamed.gif\" alt=\"\" class=\"wp-image-1553\"\/><\/figure>\n<\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Image <a href=\"https:\/\/deepmind.com\/blog\/article\/wavenet-generative-model-raw-audio\" target=\"_blank\" rel=\"noreferrer noopener\"><u>Credits<\/u><\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">In the <a href=\"https:\/\/www.theaidream.com\/post\/audio-data-analysis-using-deep-learning-with-python-part-1\" target=\"_blank\" rel=\"noreferrer noopener\"><u>previous article<\/u><\/a>, we started our discussion about audio signals; we saw how we can interpret and visualize them using Librosa python library. We also learned how to extract necessary features from a sound\/audio file.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">We concluded the previous article by building an Artificial Neural Network(ANN) for the music genre classification.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">In this article, we are going to build a Convolutional Neural Network for music genre classification.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Nowadays, deep learning is more and more used for Music Genre Classification: particularly Convolutional Neural Networks (CNN) taking as entry a spectrogram considered as an image on which are sought different types of structure.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Convolutional Neural Networks (CNN) are very similar to ordinary Neural Networks: they are made up of neurons that have learnable weights and biases. Each neuron receives some inputs, performs a dot product, and optionally follows it with a non-linearity. The whole network still expresses a single differentiable score function: from the raw image pixels on one end to class scores at the other. And they still have a loss function (e.g. SVM\/Softmax) on the last (fully-connected) layer and all the tips\/tricks we developed for learning regular Neural Networks still apply.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">So what changes? ConvNet architectures make the explicit assumption that the inputs are images, which allows us to encode certain properties into the architecture. These then make the forward function more efficient to implement and vastly reduce the number of parameters in the network.<\/p>\n\n\n\n<figure class=\"wp-block-image\"><img decoding=\"async\" src=\"https:\/\/static.wixstatic.com\/media\/3eee0b_1a12137b1cff41bcad7b85aee90596cc~mv2.gif\" alt=\"\"\/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/gfycat.com\/deadlydeafeningatlanticblackgoby-three-blue-one-brown-machines-learning\" target=\"_blank\" rel=\"noreferrer noopener\"><u>Source<\/u><\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">They are capable of detecting primary features, which are then combined by subsequent layers of the CNN architecture, resulting in the detection of higher-order complex and relevant novel features.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The dataset consists of 1000 audio tracks each 30 seconds long. It contains 10 genres, each represented by 100 tracks. The tracks are all 22050 Hz monophonic 16-bit audio files in .wav format.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The dataset can be download from <a href=\"http:\/\/marsyas.info\/downloads\/datasets.html\" target=\"_blank\" rel=\"noreferrer noopener\"><u>marsyas website.<\/u><\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">It consists of 10 genres i.e<\/p>\n\n\n\n<ul class=\"wp-block-list\"><li>Blues<\/li><li>Classical<\/li><li>Country<\/li><li>Disco<\/li><li>Hip-hop<\/li><li>Jazz<\/li><li>Metal<\/li><li>Pop<\/li><li>Reggae<\/li><li>Rock<\/li><\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Each genre contains 100 songs. Total dataset: 1000 songs.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Before moving ahead, I would recommend using <a href=\"https:\/\/colab.research.google.com\/notebooks\/intro.ipynb#recent=true\" target=\"_blank\" rel=\"noreferrer noopener\"><u>Google Colab<\/u><\/a> for doing everything related to Neural networks because it is <strong>free<\/strong> and provides GPUs and TPUs as runtime environments.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"viewer-1kr94\">Convolutional Neural Network implementation<\/h2>\n\n\n\n<figure class=\"wp-block-gallery has-nested-images columns-default is-cropped wp-block-gallery-2 is-layout-flex wp-block-gallery-is-layout-flex\">\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" width=\"1024\" height=\"385\" data-id=\"1554\" src=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125214-1024x385.jpg\" alt=\"\" class=\"wp-image-1554\" srcset=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125214-1024x385.jpg 1024w, https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125214-300x113.jpg 300w, https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125214-768x289.jpg 768w, https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125214.jpg 1042w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Steps involved in extracting features from an input spectrogram and loading it to a fully connected layer.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">So let us start building a CNN for genre classification.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">To start with load all the required libraries.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">import pandas as pd\nimport numpy as np\nfrom numpy import argmax\nimport matplotlib.pyplot as plt\n%matplotlib inline\nimport librosa\nimport librosa.display\nimport IPython.display\nimport random\nimport warnings\nimport os\nfrom PIL import Image\nimport pathlib\nimport csv\n# sklearn Preprocessing\nfrom sklearn.model_selection import train_test_split\n#Keras\nimport keras\nimport warnings\nwarnings.filterwarnings('ignore')\nfrom keras import layers\nfrom keras.layers import Activation, Dense, Dropout, Conv2D, Flatten, MaxPooling2D, GlobalMaxPooling2D, GlobalAveragePooling1D, AveragePooling2D, Input, Add\nfrom keras.models import Sequential\nfrom keras.optimizers import SGD<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">Now convert the audio data files into PNG format images or basically extracting the Spectrogram for every Audio. We will use librosa python library to extract Spectrogram for every audio file.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">genres = 'blues classical country disco hiphop jazz metal pop reggae rock'.split()\nfor g in genres:\n    pathlib.Path(f'img_data\/{g}').mkdir(parents=True, exist_ok=True)\n    for filename in os.listdir(f'.\/drive\/My Drive\/genres\/{g}'):\n        songname = f'.\/drive\/My Drive\/genres\/{g}\/{filename}'\n        y, sr = librosa.load(songname, mono=True, duration=5)\n        print(y.shape)\n        plt.specgram(y, NFFT=2048, Fs=2, Fc=0, noverlap=128, cmap=cmap, sides='default', mode='default', scale='dB');\n        plt.axis('off');\n        plt.savefig(f'img_data\/{g}\/{filename[:-3].replace(\".\", \"\")}.png')\n        plt.clf()<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">The above code will create a directory <strong>img_data <\/strong>containing all the images categorized in the genre.<\/p>\n\n\n\n<figure class=\"wp-block-gallery has-nested-images columns-default is-cropped wp-block-gallery-3 is-layout-flex wp-block-gallery-is-layout-flex\">\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" width=\"542\" height=\"737\" data-id=\"1555\" src=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125431.jpg\" alt=\"\" class=\"wp-image-1555\" srcset=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125431.jpg 542w, https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125431-221x300.jpg 221w\" sizes=\"(max-width: 542px) 100vw, 542px\" \/><\/figure>\n<\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Sample spectrograms of Disco, Classical, Blues, and Country genres respectively.<a><\/a><a><\/a><\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"565\" height=\"257\" src=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125623.jpg\" alt=\"\" class=\"wp-image-1556\" srcset=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125623.jpg 565w, https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125623-300x136.jpg 300w\" sizes=\"(max-width: 565px) 100vw, 565px\" \/><figcaption>Disco and Classical<\/figcaption><\/figure>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"562\" height=\"242\" src=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125715.jpg\" alt=\"\" class=\"wp-image-1557\" srcset=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125715.jpg 562w, https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125715-300x129.jpg 300w\" sizes=\"(max-width: 562px) 100vw, 562px\" \/><figcaption>Country and Blues<\/figcaption><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Our next step is to split the data into the train set and test set.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Install split_folders.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">pip install split_folders<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">We will split data by 80% in training and 20% in the test set.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">import split_folders\n# To only split into training and validation set, set a tuple to `ratio`, i.e, `(.8, .2)`.\nsplit_folders.ratio('.\/img_data\/', output=\".\/data\", seed=1337, ratio=(.8, .2)) # default values<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">The above code returns 2 directories for train and test set inside a parent directory.<\/p>\n\n\n\n<figure class=\"wp-block-gallery has-nested-images columns-default is-cropped wp-block-gallery-4 is-layout-flex wp-block-gallery-is-layout-flex\">\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"498\" height=\"725\" data-id=\"1558\" src=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125828.jpg\" alt=\"\" class=\"wp-image-1558\" srcset=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125828.jpg 498w, https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-125828-206x300.jpg 206w\" sizes=\"(max-width: 498px) 100vw, 498px\" \/><\/figure>\n<\/figure>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"viewer-87pah\">Image Augmentation<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Image Augmentation artificially creates training images through different ways of processing or a combination of multiple processing, such as random rotation, shifts, shear and flips, etc.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Perform Image Augmentation instead of training your model with lots of images we can train our model with fewer images and training the model with different angles and modifying the images.<\/p>\n\n\n\n<figure class=\"wp-block-gallery has-nested-images columns-default is-cropped wp-block-gallery-5 is-layout-flex wp-block-gallery-is-layout-flex\">\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"605\" height=\"350\" data-id=\"1559\" src=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/0_Utma-dS47hSoQ6Zt.png\" alt=\"\" class=\"wp-image-1559\" srcset=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/0_Utma-dS47hSoQ6Zt.png 605w, https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/0_Utma-dS47hSoQ6Zt-300x174.png 300w\" sizes=\"(max-width: 605px) 100vw, 605px\" \/><\/figure>\n<\/figure>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/towardsdatascience.com\/machinex-image-data-augmentation-using-keras-b459ef87cd22\" target=\"_blank\" rel=\"noreferrer noopener\"><u>Image Augmentation<\/u><\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Keras has this <strong>ImageDataGenerator<\/strong> class which allows the users to perform image augmentation on the fly in a very easy way. You can read about that in Keras\u2019s official <a href=\"https:\/\/keras.io\/preprocessing\/image\/\" target=\"_blank\" rel=\"noreferrer noopener\"><u>documentation<\/u><\/a>.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">from keras.preprocessing.image import ImageDataGenerator\ntrain_datagen = ImageDataGenerator(\n        rescale=1.\/255, # rescale all pixel values from 0-255, so aftre this step all our pixel values are in range (0,1)\n        shear_range=0.2, #to apply some random tranfromations\n        zoom_range=0.2, #to apply zoom\n        horizontal_flip=True) # image will be flipper horiztest_datagen = ImageDataGenerator(rescale=1.\/255)<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">The ImageDataGenerator class has three methods <strong>flow(),<\/strong> <strong>flow_from_directory() and flow_from_dataframe() <\/strong>to read the images from a big numpy array and folders containing images.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">We will discuss only flow_from_directory() in this blog post.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">training_set = train_datagen.flow_from_directory(\n        '.\/data\/train',\n        target_size=(64, 64),\n        batch_size=32,\n        class_mode='categorical',\n        shuffle = False)test_set = test_datagen.flow_from_directory(\n        '.\/data\/val',\n        target_size=(64, 64),\n        batch_size=32,\n        class_mode='categorical',\n        shuffle = False )<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>flow_from_directory()<\/strong> has the following arguments.<\/p>\n\n\n\n<ul class=\"wp-block-list\"><li><strong>directory: <\/strong>path where there exists a folder, under which all the test images are present. For example, in this case, the training images are found in .\/data\/train<\/li><li><strong>batch_size: <\/strong>Set this to some number that divides your total number of images in your test set exactly. <strong>Why this only for test_generator?<\/strong> Actually, you should set the \u201cbatch_size\u201d in both train and valid generators to some number that divides your total number of images in your train set and valid respectively, but this doesn\u2019t matter before because even if batch_size doesn\u2019t match the number of samples in the train or valid sets and some images gets missed out every time we yield the images from generator, it would be sampled the very next epoch you train. But for the test set, you should sample the images exactly once, no less or no more. If Confusing, just set it to 1(but maybe a little bit slower).<\/li><li><strong>class_mode:<\/strong> Set \u201cbinary\u201d if you have only two classes to predict, if not set to\u201ccategorical\u201d, in case if you\u2019re developing an Autoencoder system, both input and the output would probably be the same image, for this case set to \u201cinput\u201d.<\/li><li><strong>shuffle:<\/strong> Set this to <em>False,<\/em> because you need to yield the images in \u201corder\u201d, to predict the outputs and match them with their unique ids or filenames.<\/li><\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Create a Convolutional Neural Network:<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">model = Sequential()\ninput_shape=(64, 64, 3)#1st hidden layer\nmodel.add(Conv2D(32, (3, 3), strides=(2, 2), input_shape=input_shape))\nmodel.add(AveragePooling2D((2, 2), strides=(2,2)))\nmodel.add(Activation('relu'))#2nd hidden layer\nmodel.add(Conv2D(64, (3, 3), padding=\"same\"))\nmodel.add(AveragePooling2D((2, 2), strides=(2,2)))\nmodel.add(Activation('relu'))#3rd hidden layer\nmodel.add(Conv2D(64, (3, 3), padding=\"same\"))\nmodel.add(AveragePooling2D((2, 2), strides=(2,2)))\nmodel.add(Activation('relu'))#Flatten\nmodel.add(Flatten())\nmodel.add(Dropout(rate=0.5))#Add fully connected layer.\nmodel.add(Dense(64))\nmodel.add(Activation('relu'))\nmodel.add(Dropout(rate=0.5))#Output layer\nmodel.add(Dense(10))\nmodel.add(Activation('softmax'))model.summary()<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">Compile\/train the network using <a href=\"https:\/\/en.wikipedia.org\/wiki\/Stochastic_gradient_descent\" target=\"_blank\" rel=\"noreferrer noopener\"><u>Stochastic Gradient Descent(SGD)<\/u><\/a>. Gradient Descent works fine when we have a convex curve. But if we don\u2019t have a convex curve, Gradient Descent fails. Hence, in Stochastic Gradient Descent, few samples are selected randomly instead of the whole data set for each iteration.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">epochs = 200\nbatch_size = 8\nlearning_rate = 0.01\ndecay_rate = learning_rate \/ epochs\nmomentum = 0.9\nsgd = SGD(lr=learning_rate, momentum=momentum, decay=decay_rate, nesterov=False)\nmodel.compile(optimizer=\"sgd\", loss=\"categorical_crossentropy\", metrics=['accuracy'])<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">Now fit the model with 50 epochs.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">model.fit_generator(\n        training_set,\n        steps_per_epoch=100,\n        epochs=50,\n        validation_data=test_set,\n        validation_steps=200)<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">Now since the CNN model is trained, let us evaluate it. evaluate_generator() uses both your test input and output. It first predicts output using training input and then evaluates the performance by comparing it against your test output. So it gives out a measure of performance, i.e. accuracy in your case.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">#Model Evaluation\nmodel.evaluate_generator(generator=test_set, steps=50)#OUTPUT\n[1.704445120342617, 0.33798882681564246]<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">So the loss is 1.70 and Accuracy is 33.7%.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">At last, let your model make some predictions on the test data set. You need to reset the test_set before whenever you call the <strong>predict_generator<\/strong>. This is important, if you forget to reset the <strong>test_set<\/strong> you will get outputs in a weird order.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">test_set.reset()\npred = model.predict_generator(test_set, steps=50, verbose=1)<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">As of now <strong>predicted_class_indices<\/strong> has the predicted labels, but you can\u2019t simply tell what the predictions are, because all you can see is numbers like 0,1,4,1,0,6\u2026 You need to map the predicted labels with their unique ids such as filenames to find out what you predicted for which image.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">predicted_class_indices=np.argmax(pred,axis=1)\n\nlabels = (training_set.class_indices)\nlabels = dict((v,k) for k,v in labels.items())\npredictions = [labels[k] for k in predicted_class_indices]\npredictions = predictions[:200]\nfilenames=test_set.filenames<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">Append <strong><em>filenames<\/em><\/strong> and <strong><em>predictions<\/em><\/strong> to a single pandas data frame as two separate columns. But before doing that check the size of both, it should be the same.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">print(len(filename, len(predictions)))\n# (200, 200)<\/pre>\n\n\n\n<p class=\"wp-block-paragraph\">Finally, save the results to a CSV file.<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">results=pd.DataFrame({\"Filename\":filenames,\n                      \"Predictions\":predictions},orient='index')\nresults.to_csv(\"prediction_results.csv\",index=False)<\/pre>\n\n\n\n<figure class=\"wp-block-gallery has-nested-images columns-default is-cropped wp-block-gallery-6 is-layout-flex wp-block-gallery-is-layout-flex\">\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"832\" height=\"872\" data-id=\"1560\" src=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-130045.jpg\" alt=\"\" class=\"wp-image-1560\" srcset=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-130045.jpg 832w, https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-130045-286x300.jpg 286w, https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/Screenshot-2021-04-06-130045-768x805.jpg 768w\" sizes=\"(max-width: 832px) 100vw, 832px\" \/><\/figure>\n<\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Output<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">I have trained the model on 50 epochs(which itself took 1.5 hours to execute on Nvidia K80 GPU). If you wanna increase the accuracy, increase the number of epochs to 1000 or even more while training your CNN model.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">You can also find code on my Github: <a href=\"https:\/\/github.com\/nageshsinghc4\/Audio-Data-Analysis-Using-Deep-Learning\" target=\"_blank\" rel=\"noreferrer noopener\"><u>https:\/\/github.com\/nageshsinghc4\/Audio-Data-Analysis-Using-Deep-Learning<\/u><\/a><\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"viewer-1ldab\">Conclusion<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">So it shows that CNN is a viable alternative for automatic feature extraction. Such discovery lends support to our hypothesis that the intrinsic characteristics in the variation of musical data are similar to those of image data. Our CNN model is highly scalable but not robust enough to generalized the training result to unseen musical data. This can be overcome with an enlarged dataset and of course the amount of dataset that can be fed.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Well, this concludes the two-article series on Audio Data Analysis Using Deep Learning with Python. I hope you guys have enjoyed reading it, feel free to share your comments\/thoughts\/feedback in the comment section.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"viewer-896ag\">Thanks for reading this article!!!<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\"><em><strong>Note:\u00a0<\/strong>T<em>his article was originally published on\u00a0<a href=\"https:\/\/www.theaidream.com\/post\/audio-data-analysis-using-deep-learning-with-python-part-2\">theaidream<\/a>, and kindly contributed to AI Planet (formerly DPhi) to spread the knowledge.<\/em><\/em><\/p>\n","protected":false},"excerpt":{"rendered":"<p>If you are an absolute beginner to audio data analysis, you should read Part 1 of this series first. Once you are comfortable with the concepts explained in that article, you can come back and continue with this article. Image Credits In the previous article, we started our discussion about audio signals; we saw how [&hellip;]<\/p>\n","protected":false},"author":23,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[68,20,69,22],"class_list":["post-1552","post","type-post","status-publish","format-standard","hentry","category-uncategorized","tag-audio-analysis","tag-datascience","tag-deep-learning","tag-machine-learning"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.8 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Audio Data Analysis Using Deep Learning with Python (Part 2) - AI Planet<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Audio Data Analysis Using Deep Learning with Python (Part 2) - AI Planet\" \/>\n<meta property=\"og:description\" content=\"If you are an absolute beginner to audio data analysis, you should read Part 1 of this series first. Once you are comfortable with the concepts explained in that article, you can come back and continue with this article. Image Credits In the previous article, we started our discussion about audio signals; we saw how [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/\" \/>\n<meta property=\"og:site_name\" content=\"AI Planet\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/aiplanet\/\" \/>\n<meta property=\"article:published_time\" content=\"2021-04-06T07:31:35+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2022-11-15T09:27:54+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/unnamed.gif\" \/>\n<meta name=\"author\" content=\"Nagesh Singh Chauhan\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Nagesh Singh Chauhan\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"9 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/\"},\"author\":{\"name\":\"Nagesh Singh Chauhan\",\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/#\\\/schema\\\/person\\\/f7f7394ec66e5136b2b07cc29ebf1fb8\"},\"headline\":\"Audio Data Analysis Using Deep Learning with Python (Part 2)\",\"datePublished\":\"2021-04-06T07:31:35+00:00\",\"dateModified\":\"2022-11-15T09:27:54+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/\"},\"wordCount\":1362,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/wp-content\\\/uploads\\\/2021\\\/04\\\/unnamed.gif\",\"keywords\":[\"audio analysis\",\"datascience\",\"deep learning\",\"machine learning\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/\",\"url\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/\",\"name\":\"Audio Data Analysis Using Deep Learning with Python (Part 2) - AI Planet\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/wp-content\\\/uploads\\\/2021\\\/04\\\/unnamed.gif\",\"datePublished\":\"2021-04-06T07:31:35+00:00\",\"dateModified\":\"2022-11-15T09:27:54+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/#primaryimage\",\"url\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/wp-content\\\/uploads\\\/2021\\\/04\\\/unnamed.gif\",\"contentUrl\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/wp-content\\\/uploads\\\/2021\\\/04\\\/unnamed.gif\",\"width\":570,\"height\":289},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/audio-data-analysis-using-deep-learning-with-python-part-2\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Audio Data Analysis Using Deep Learning with Python (Part 2)\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/#website\",\"url\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/\",\"name\":\"AI Planet, Formerly DPhi\",\"description\":\"AI Planet (formerly DPhi) is a global ecosystem that aims to educate and empower individuals and organizations to build and utilize artificial intelligence (AI). With a community of over 300,000 AI specialists and enthusiasts, we offer a range of products and have helped drive innovative solutions for various industries. Our ultimate goal is to make AI education accessible to everyone and use AI to tackle important business and social challenges.\",\"publisher\":{\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/#organization\",\"name\":\"AI Planet, Formerly DPhi\",\"url\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/wp-content\\\/uploads\\\/2022\\\/11\\\/Size48-Large-1.png\",\"contentUrl\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/wp-content\\\/uploads\\\/2022\\\/11\\\/Size48-Large-1.png\",\"width\":48,\"height\":48,\"caption\":\"AI Planet, Formerly DPhi\"},\"image\":{\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/www.facebook.com\\\/aiplanet\\\/\",\"https:\\\/\\\/www.instagram.com\\\/aiplanet_community\\\/\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/aiplanet\\\/\",\"https:\\\/\\\/www.youtube.com\\\/@aiplanet\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/#\\\/schema\\\/person\\\/f7f7394ec66e5136b2b07cc29ebf1fb8\",\"name\":\"Nagesh Singh Chauhan\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/b8a063fd73beeb4a06ec6a22acb32b5ceeb60ee5896f3583d09eaf03ffc7537b?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/b8a063fd73beeb4a06ec6a22acb32b5ceeb60ee5896f3583d09eaf03ffc7537b?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/b8a063fd73beeb4a06ec6a22acb32b5ceeb60ee5896f3583d09eaf03ffc7537b?s=96&d=mm&r=g\",\"caption\":\"Nagesh Singh Chauhan\"},\"description\":\"Nagesh is Medium's Official Top writer in Artificial Intelligence. He loves to share knowledge through his articles and comes with rich experience across various domains related to Data Science, Machine learning, Neural networks, Text Analysis, NLP and others.\",\"url\":\"https:\\\/\\\/aiplanet.com\\\/blog\\\/author\\\/nagesh\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Audio Data Analysis Using Deep Learning with Python (Part 2) - AI Planet","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/","og_locale":"en_US","og_type":"article","og_title":"Audio Data Analysis Using Deep Learning with Python (Part 2) - AI Planet","og_description":"If you are an absolute beginner to audio data analysis, you should read Part 1 of this series first. Once you are comfortable with the concepts explained in that article, you can come back and continue with this article. Image Credits In the previous article, we started our discussion about audio signals; we saw how [&hellip;]","og_url":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/","og_site_name":"AI Planet","article_publisher":"https:\/\/www.facebook.com\/aiplanet\/","article_published_time":"2021-04-06T07:31:35+00:00","article_modified_time":"2022-11-15T09:27:54+00:00","og_image":[{"url":"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/unnamed.gif","type":"","width":"","height":""}],"author":"Nagesh Singh Chauhan","twitter_card":"summary_large_image","twitter_misc":{"Written by":"Nagesh Singh Chauhan","Est. reading time":"9 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/#article","isPartOf":{"@id":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/"},"author":{"name":"Nagesh Singh Chauhan","@id":"https:\/\/aiplanet.com\/blog\/#\/schema\/person\/f7f7394ec66e5136b2b07cc29ebf1fb8"},"headline":"Audio Data Analysis Using Deep Learning with Python (Part 2)","datePublished":"2021-04-06T07:31:35+00:00","dateModified":"2022-11-15T09:27:54+00:00","mainEntityOfPage":{"@id":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/"},"wordCount":1362,"commentCount":0,"publisher":{"@id":"https:\/\/aiplanet.com\/blog\/#organization"},"image":{"@id":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/#primaryimage"},"thumbnailUrl":"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/unnamed.gif","keywords":["audio analysis","datascience","deep learning","machine learning"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/","url":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/","name":"Audio Data Analysis Using Deep Learning with Python (Part 2) - AI Planet","isPartOf":{"@id":"https:\/\/aiplanet.com\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/#primaryimage"},"image":{"@id":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/#primaryimage"},"thumbnailUrl":"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/unnamed.gif","datePublished":"2021-04-06T07:31:35+00:00","dateModified":"2022-11-15T09:27:54+00:00","breadcrumb":{"@id":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/#primaryimage","url":"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/unnamed.gif","contentUrl":"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2021\/04\/unnamed.gif","width":570,"height":289},{"@type":"BreadcrumbList","@id":"https:\/\/aiplanet.com\/blog\/audio-data-analysis-using-deep-learning-with-python-part-2\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/aiplanet.com\/blog\/"},{"@type":"ListItem","position":2,"name":"Audio Data Analysis Using Deep Learning with Python (Part 2)"}]},{"@type":"WebSite","@id":"https:\/\/aiplanet.com\/blog\/#website","url":"https:\/\/aiplanet.com\/blog\/","name":"AI Planet, Formerly DPhi","description":"AI Planet (formerly DPhi) is a global ecosystem that aims to educate and empower individuals and organizations to build and utilize artificial intelligence (AI). With a community of over 300,000 AI specialists and enthusiasts, we offer a range of products and have helped drive innovative solutions for various industries. Our ultimate goal is to make AI education accessible to everyone and use AI to tackle important business and social challenges.","publisher":{"@id":"https:\/\/aiplanet.com\/blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/aiplanet.com\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/aiplanet.com\/blog\/#organization","name":"AI Planet, Formerly DPhi","url":"https:\/\/aiplanet.com\/blog\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/aiplanet.com\/blog\/#\/schema\/logo\/image\/","url":"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2022\/11\/Size48-Large-1.png","contentUrl":"https:\/\/aiplanet.com\/blog\/wp-content\/uploads\/2022\/11\/Size48-Large-1.png","width":48,"height":48,"caption":"AI Planet, Formerly DPhi"},"image":{"@id":"https:\/\/aiplanet.com\/blog\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/aiplanet\/","https:\/\/www.instagram.com\/aiplanet_community\/","https:\/\/www.linkedin.com\/company\/aiplanet\/","https:\/\/www.youtube.com\/@aiplanet"]},{"@type":"Person","@id":"https:\/\/aiplanet.com\/blog\/#\/schema\/person\/f7f7394ec66e5136b2b07cc29ebf1fb8","name":"Nagesh Singh Chauhan","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/b8a063fd73beeb4a06ec6a22acb32b5ceeb60ee5896f3583d09eaf03ffc7537b?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/b8a063fd73beeb4a06ec6a22acb32b5ceeb60ee5896f3583d09eaf03ffc7537b?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/b8a063fd73beeb4a06ec6a22acb32b5ceeb60ee5896f3583d09eaf03ffc7537b?s=96&d=mm&r=g","caption":"Nagesh Singh Chauhan"},"description":"Nagesh is Medium's Official Top writer in Artificial Intelligence. He loves to share knowledge through his articles and comes with rich experience across various domains related to Data Science, Machine learning, Neural networks, Text Analysis, NLP and others.","url":"https:\/\/aiplanet.com\/blog\/author\/nagesh\/"}]}},"_links":{"self":[{"href":"https:\/\/aiplanet.com\/blog\/wp-json\/wp\/v2\/posts\/1552","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/aiplanet.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/aiplanet.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/aiplanet.com\/blog\/wp-json\/wp\/v2\/users\/23"}],"replies":[{"embeddable":true,"href":"https:\/\/aiplanet.com\/blog\/wp-json\/wp\/v2\/comments?post=1552"}],"version-history":[{"count":4,"href":"https:\/\/aiplanet.com\/blog\/wp-json\/wp\/v2\/posts\/1552\/revisions"}],"predecessor-version":[{"id":4911,"href":"https:\/\/aiplanet.com\/blog\/wp-json\/wp\/v2\/posts\/1552\/revisions\/4911"}],"wp:attachment":[{"href":"https:\/\/aiplanet.com\/blog\/wp-json\/wp\/v2\/media?parent=1552"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/aiplanet.com\/blog\/wp-json\/wp\/v2\/categories?post=1552"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/aiplanet.com\/blog\/wp-json\/wp\/v2\/tags?post=1552"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}