Editing Deep Learning Revolution (section)

=== Convolutional Neural Networks (CNNs) ===
One of the most influential network architectures in computer vision is the Convolutional Neural Network (CNN). While modern CNNs can trace their origins to Yann LeCun’s LeNet-5, initially designed for digit recognition, versions like AlexNet and ResNet have evolved to be more sophisticated, achieving significant success in image classification challenges.<ref name=":02" /> Although these CNNs are primarily associated with image processing, they also find great success in use of speech recognition systems, namely spectrograms of audio data. Spectrograms are 2D visual representations of the audio's frequency content over time, which are images that can be processed by CNNs. CNNs can extract data from these spectrograms, such as frequency patterns, phonemes, and other acoustic features.<ref>Abdel-Hamid, O., Mohamed, A. R., Jiang, H., & Penn, G. (2012, March). Applying convolutional neural networks concepts to hybrid NN-HMM model for speech recognition. In ''2012 IEEE international conference on Acoustics, speech and signal processing (ICASSP)'' (pp. 4277-4280). IEEE.</ref><ref>Abdel-Hamid, O., Mohamed, A. R., Jiang, H., Deng, L., Penn, G., & Yu, D. (2014). Convolutional neural networks for speech recognition. ''IEEE/ACM Transactions on audio, speech, and language processing'', ''22''(10), 1533-1545.</ref>