Skip to content

Acustic events classifictaion using CNN on UrbanSound8K dataset

Notifications You must be signed in to change notification settings

GKPr0/AcusticEventsDetection

Repository files navigation

Urban Sound classification using CNN

Dataset

Data analysis

  • Analysis of dataset such as recording lengths, class distribution or sampling frequencies.
  • Listening to class samples.
  • Visualization of data as waveform, STFT, MEL-STFT, MFCC.
  • Jupyter notebook
  • Google colab

CNN Training

  • Trying to classify based on MelSpectogram.
  • Data augmented with time shifting, time masking and freq masking.
  • Model is saved after each epoch.
  • Confusion matrix to peak at model behavior.
  • Jupyter notebook
  • Google colab

Environment

  • Python 3.8 (recommend to create virtual env)

      python38 -m venv .venv
      source .venv/Scripts/Activate
    
  • All necessary dependencies in requirements.txt

      pip install -r requirements.txt
    

About

Acustic events classifictaion using CNN on UrbanSound8K dataset

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published