Skip to content

Extract audio features from .wav files and classify genres using TensorFlow

Notifications You must be signed in to change notification settings

vladinator1000/tfgenre

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

tfgenre - Vlady Veselinov

Download and unzip trained models in root folder (2GB)

To evaluate accuracy of all models:

sh evaluate.sh

To train all models

sh train.sh

Optional audio data extraction guide with Essentia (macOS only)

Data set used for this project can be found here

  1. Install Essentia with Homebrew: brew tap MTG/essentia && brew install essentia

  2. Make a data/audio folder in the project root containing subfolders of audio files. Subfolder names used as categories, i.e. data/audio/car_horn/11251.wav

  3. Run src/extract_features.py with Python2.7 to load audio from data/audio and produce JSON files with audio features

  4. Run write_csv.py to distill all JSON to one .csv

Optional Run TensorBoard for visualisations:

Download and unzip log files in root folder

Start TensorBoard

sh tensorboard.sh

About

Extract audio features from .wav files and classify genres using TensorFlow

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published