FEATURE REQUESTED : Urban Sound Analysis - Sound Classification

[SimranShaikh20]

This project classifies audio samples from urban environments into one of 10 classes. The dataset, known as UrbanSound8K, contains 8732 sound excerpts, each 4 seconds or shorter, representing urban sounds like air conditioners, car horns, children playing, and more. Using a Neural Network model, the project aims to achieve high classification accuracy (reported at 93%).

Key Objectives : Classify Urban Sound Events , Feature Extraction from Audio Data , Use Deep Learning for Accuracy

Overall Goal:
To build a robust and accurate audio classification model that can recognize urban sound types, offering potential for real-world applications in smart cities and environmental monitoring systems.

@abhisheks008 liked my idea / new feature !
then assign me this issue to work on it .

[github-actions]

Thank you for creating this issue! We'll look into it as soon as possible. Your contributions are highly appreciated! 😊

[abhisheks008]

Hi @SimranShaikh20 what are the models you are planning to implement here?
Can you please share the dataset URL?

[SimranShaikh20]

here is dataset link - https://drive.google.com/file/d/1EpRQYWDIiZgmgnQunOFCzGIrfissXOxX/view?usp=sharing
i am thinking to implement it by using 3 ways
and model i will implemet is - Training a ResNet model on spectorgram images / Training a Convolutional neural network on spectrogram images / Training an Artificial Neural Network on time-series audio data

but i will select more prefereable which is ann
you can suggest me which i can used from above 3 option

@abhisheks008

[SimranShaikh20]

hey @abhisheks008 can you assign me this project as i ha shared all mandatory details !

[SimranShaikh20]

@abhisheks008 let me know have you liked my idea !

[abhisheks008]

Hi @SimranShaikh20 try to implement in all the three possible ways mentioned by you, then based on the accuracy scores you can figure out the best fitted approach.

Assigned to you.