Paper Title
Audio Pattern Recognition using Convolutional Neural Network

Abstract
Recently enormous work is done in the field of audio. In the Machine learning area, Audio Pattern recognition (APR) is an important topic with multiple tasks such as sound event detection, music classification, Audio classification, Acoustic scene classification, Speechemotion classification, and audio tagging. In this paper, we are presenting the training-based approach for Audio Pattern Recognition using CNN (Convolutional Neural Networks). There are different data sets available for this task, but they are having some disadvantages like a limitation for the duration [1] or having some parameters. For this, we created a custom dataset from different videos and audios. A different audio source is collected from some online sources whereas some are collected from the recording. In this dataset, there are audios available from 3 seconds to 5 seconds. In this paper, training and evaluating the data for recognizing patterns of audio for 10 classes is done. From the result of the training, it can be said that this dataset is useful for the different tasks which are used for audio Keywords - Audio Pattern Recognition, Convolutional Neural Network, Dataset.