jaddoughman / Audio-Data-Pre-Processing

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Audio-Data-Pre-Processing

Part 1: Data Mining

You are given a soundscloud_urls.csv file containing URLs to SoundCloud.com. You are tasked with extracting the audio files, the name of the author and the track name from the provided links, save this data in a .csv and display the dataframe in your notebook. All audio file must be saved as .wav in the ./data directory.

Part 2: Data Processing and Visualization

Once you have downloaded your audio files, you are tasked with pre-processing your data by converting the raw waveform to a spectrogram using the Fast Fourier Transform (FFT). You can find more information about the FFT HERE Finally, your must convert your spectrogram from Hz range to the mel-scale. More information about me-scaling can be found HERE

Note: You are not permitted to use built-in functions to perform FFT and mel-scale operations mentioned in part 2.

About


Languages

Language:Jupyter Notebook 100.0%