Extracting Mel Spectrograms with Pytorch and Torchaudio
Valerio Velardo - The Sound of AI via YouTube
Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore the process of extracting Mel spectrograms and resampling audio using PyTorch and torchaudio in this comprehensive 23-minute tutorial. Dive into the most common torchaudio transforms and learn how to apply them effectively. Follow along as the instructor demonstrates instantiating MelSpectrogram, extracting Mel spectrograms from the UrbanSoundDataset, and implementing resampling and mixing down techniques in the getitem method. Gain practical insights into resampling signals, converting audio to mono, and running scripts to extract Mel spectrograms. Access the accompanying code on GitHub to enhance your understanding and practice the concepts covered in this informative video.
Syllabus
Intro
Torchaudio transformations
Instantiating MelSpectrogram
Extracting Mel spectrograms in UrbanSoundDataset
Resample and mix down in getitem
Resampling signal
Mixing down signal to mono
Getitem recap
Running the script to extract mel spectrogram
Outro
Taught by
Valerio Velardo - The Sound of AI