audio-classification

torchaudio and associated libraries to build a DNN for audio classification

Tackling the problem as a Computer Vision and Image Processing problem

We tackle the problem as vision problem rather than working in 1D waveform. We create mel spectrograms from the audio and perform the classification of audios

Using the Urban8K dataset for classification. Link

Excellent resource for leanring audio signal processing can be found here Link