Deepfake-Audio-Recognition
Background
The advent of machine learning led to increased innovation. In most cases this is a good thing however, there are people who use this for their own benefit and in the process harm others.
This project is centered on identifying the difference between ones actual voice and a trained machine learning voice. With the use of voice recognition to access banking services or to authenticate oneself on the rise, we would like to create a system that will be able to identify deepfakes and avert the risk of people being able to access each other’s personal information. The project will be an extension of Jemine, Corentin works, who provided a vocoder that is able to clone human speech in realtime.
Objectives
The main objective will be to detect the difference between real and computer generated voices and correctly classify them.
Specific Objectives:
- To create a webapp for audio classification.
- To identify characteristics of deep fakes vs those of natural audio.
Built With
We would love to continue improving this analysis. Please contribute..
Contributions are what make the open source community such an amazing place to be learn, inspire, and create. Any contributions you make are greatly appreciated.
- Fork the Project
- Create your Feature Branch
- Commit your Changes
- Push to the Branch
- Open a Pull Request
License
Distributed under the MIT License. See LICENSE for more information.