https://github.com/mdfirman/citynet
A neural network classifier for urban soundscapes
https://github.com/mdfirman/citynet
audio-classification biodiversity deep-learning ecoacoustics ecology london neural-network soundscapes urban-ecology
Last synced: 7 months ago
JSON representation
A neural network classifier for urban soundscapes
- Host: GitHub
- URL: https://github.com/mdfirman/citynet
- Owner: mdfirman
- Created: 2015-08-21T09:17:30.000Z (almost 10 years ago)
- Default Branch: master
- Last Pushed: 2024-01-24T10:27:42.000Z (over 1 year ago)
- Last Synced: 2024-01-29T06:35:13.036Z (over 1 year ago)
- Topics: audio-classification, biodiversity, deep-learning, ecoacoustics, ecology, london, neural-network, soundscapes, urban-ecology
- Language: Jupyter Notebook
- Homepage: http://londonsounds.org/
- Size: 54.2 MB
- Stars: 27
- Watchers: 6
- Forks: 5
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# CityNet - a neural network for urban sounds
CityNet is a machine-learned system for estimating the level of **biotic** and **anthropogenic** sound at each moment in time in an audio file.
The system has been trained and validated on human-labelled audio files captured from green spaces around London.
CityNet comprises a neural network classifier, which operates on audio spectrograms to produce a measure of biotic or anthropogenic activity level.
More details of the method are available from the paper:
> **[CityNet - Deep Learning Tools for Urban Ecoacoustic Assessment](https://doi.org/10.1101/248708)**
>
> Alison J Fairbrass, Michael Firman, Carol Williams, Gabriel J Brostow, Helena Titheridge and Kate E Jones
>
> **doi**: https://doi.org/10.1101/248708
An overview of predictions of biotic and anthropogenic activity on recordings of London sounds can be seen at our website [londonsounds.org](http://londonsounds.org).
[](http://londonsounds.org)
## Requirements
The system has been tested using the dependencies in `environment.yml`. Our code works with python 3.
You can create an environment with all the dependencies installed using:
```bash
conda env create -f environment.yml -n citynet
conda activate citynet
```
## How to classify a new audio file with CityNet
- Run `python demo.py` to classify an example audio file.
- Predictions should be saved in the folder `demo`.
- Your newly-created file `demo/prediction.pdf` should look identical to the provided file `demo/reference_prediction.pdf`:
## How to classify multiple audio files
You can run CityNet on a folder of audio files with:
```bash
python multi_predict.py path/to/audio/files
```
This will save summaries of what is found in each wav file found to `prediction_summaries.csv`.
## Hardware requirements
For training and testing we used a 2GB NVIDIA GPU. The computation requirements for classification are pretty low though, so a GPU should not be required.