# Environmental-Sound-Classification-Using-Convolutional-Neural-Networks

**Repository Path**: zhouchunlei/Environmental-Sound-Classification-Using-Convolutional-Neural-Networks

## Basic Information

- **Project Name**: Environmental-Sound-Classification-Using-Convolutional-Neural-Networks
- **Description**: Jupyter notebooks and Python modules for the processing of the dataset (UrbanSound8K) and the implementation of the models for the task of sound classification using convolutional neural networks. The pdf with the final version of the thesis project can also be found in this repository. 
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 1
- **Created**: 2020-07-28
- **Last Updated**: 2023-06-20

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Single and Multi-Label Environmental Sound Classification Using Convolutional Neural Networks

A master thesis project by Santiago Álvarez-Buylla Puente, carried out at Chalmers University of Technology between January and June 2018, for the Sound and Vibration MSc.



## Content of the repository

In the repository, four Jupyter notebooks can be found:

- Dataset_processing.ipynb
- Dataset_processing_Multilabel.ipynb
- Audio_Classification_Softmax_mine_260318.ipynb
- Audio_Classification_Multilabel_mine_170518.ipynb

And three Python modules:

- santiago_data_preprocessing.py
- santiago_my_modules_v3_160418.py
- santiago_my_modules_v3_17_05_18_Multilabel.py

In the two first Jupyter notebooks, the processing of the dataset is performed and explained, both for the single and the multi-label classification tasks. These two files make use of the Python module *santiago_data_preprocessing.py*  
The two following Jupyter notebooks are the notebooks where the two models are implemented, making use of the two Python modules *santiago_my_modules_v3_160418.py* and *santiago_my_modules_v3_17_05_18_Multilabel.py*


Hope you find it useful.  
Any suggestion is always welcomed!