Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/rednafi/urban-sound-classification
Urban sound source tagging from an aggregation of four second noisy audio clips via 1D and 2D CNN (Xception)
https://github.com/rednafi/urban-sound-classification
audio-processing audio-tagging classification machine-learning mel-spectrogram sound-classification sound-classification-spectrograms sound-processing sound-synthesis urban-sound-8k urban-sound-classification
Last synced: about 11 hours ago
JSON representation
Urban sound source tagging from an aggregation of four second noisy audio clips via 1D and 2D CNN (Xception)
- Host: GitHub
- URL: https://github.com/rednafi/urban-sound-classification
- Owner: rednafi
- License: mit
- Created: 2019-05-03T10:50:31.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2023-03-24T22:34:26.000Z (almost 2 years ago)
- Last Synced: 2024-12-28T10:50:34.676Z (14 days ago)
- Topics: audio-processing, audio-tagging, classification, machine-learning, mel-spectrogram, sound-classification, sound-classification-spectrograms, sound-processing, sound-synthesis, urban-sound-8k, urban-sound-classification
- Language: Jupyter Notebook
- Homepage:
- Size: 36.9 MB
- Stars: 58
- Watchers: 4
- Forks: 15
- Open Issues: 20
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Urban Sound Classification
Urban sound source tagging from an aggregation of four second noisy audio clips via 1D and 2D CNN (Xception)
[](https://urbansounddataset.weebly.com/urbansound8k.html)
[](https://www.python.org/)
[](https://github.com/rednafi/urban-sound-classification/blob/master/LICENSE)
[](https://github.com/python/black)
## Dataset Description
The Urban Sound Classification dataset contains 8732 labeled sound excerpts (<=4s) of urban sounds from 10 classes,namely:
* Air Conditioner
* Car Horn
* Children Playing
* Dog bark
* Drilling Engine
* Idling Gun Shot
* Jackhammer
* Siren
* Street Music
The attributes of data are mapped as follows:
* **ID** – Unique ID of sound excerpt and **Class** – type of sound
## Project Organization
### Folder Structure
```
.
├── data
│ ├── img
│ │ ├── audio-features.png
│ │ ├── sound.png
│ │ └── time_freq.png
│ ├── test
│ │ └── Test
| | ├── 1.wav
| | ├── 2.wav
| | ├── .............
│ ├── test.csv
│ ├── train
│ │ └── Train
| | ├── 1.wav
| | ├── 2.wav
| | ├── ............
| |
│ └── train.csv
├── LICENSE
├── notebooks
│ ├── eda_plots
│ │ ├── amplitude_vs_time
│ │ │ ├── air_conditioner.svg
│ │ │ ├── car_horn.svg
| | | ├── ............
│ │ └── mel_spectrum
│ │ ├── air_conditioner.png
│ │ ├── car_horn.png
| | ├── ............
│ └── Exploratory Data Analysis.ipynb
├── README.md
├── requirements.txt
├── results
│ ├── acc_model_1d.png
│ ├── acc_model_2d.png
│ ├── loss_model_1d.png
│ ├── loss_model_2d.png
│ ├── pred_1d.csv
│ └── pred_2d.csv
└── src
├── test_1d.py
├── test_2d.py
├── train_1d.py
├── train_2d.py
├── utils_1d.py
└── utils_2d.py
```
### Workflow
**Exploratory Data Analysis:**
* Frequency normalization and amplitude vs time plot
* Mel spectogram plot
**Audio Tagging:**
* Normalizing the audio clips and passing them through stacks of 1D convolution layers for feature extraction. Then the usual dense layer stacks were used to do the final categorization.
* Extracting features in the form of [mel-spectogram](https://en.wikipedia.org/wiki/Mel-frequency_cepstrum) and passing them
through stacks of 2D convolution layers for additional feature pulling. Dense layer stack does the final classification. In this case, we trained an Xception model from scratch to achieve better generalization capability.
### Result
We achieved 89% validation accuracy in the second approach.
### Requirements
```
pip install -r requirements.txt
```