Songs are split into overlapping chunks Each chunk is converted to a mel spectrogram image A CNN is trained to classify spectrogram images by song ID Prediction runs ...

This project exists because of a $300 mistake. While building a 4G LTE network on a BladeRF using srsRAN, corrupted packets from the radio cost five weeks of debugging with no clear path to the root ...

Abstract: In this work, we propose CleanMel, a single-channel Mel-spectrogram denoising and dereverberation network for improving both speech quality and automatic speech recognition (ASR) performance ...

Abstract: Spectrum sensing is one of the key problems in the cognitive radio network. Existing spectrum sensing methods commonly use deep learning models such as the convolutional neural network (CNN) ...