Sigma tech

• 

  group photo

Inspiration

Many initiatives regarding AI and deep fake technologies have been launched by governments around the globe. These initiatives were a response to the many unintended effects of AI usage. We wanted to make impactful use of AI technology in the local context to ensure that Singaporeans are not affected by the growing danger of deep fake content online.

What it does

Chrome extension that can detect deep fake content and warn the user about the content they are viewing.

How we built it

2 deep fake detectors run in parallel - one for audio and one for video. Using open-source machine learning models, we trained these models using the training dataset that we made. These backend codes were written in Python.

Challenges we ran into

With the time and processing power constraints, we were not able to train using the entire dataset and had to reduce the number of data to train the models.

Accomplishments that we're proud of

We were proud that, even with these constraints, we were able to come up with a full-featured solution that can be used and implemented with a bit of tweaking.

What we learned

We've learned to work under constraints by staying true to our project scope and making use of readily available open-source materials for social good.

What's next for Sigma tech

We aim to polish our solution so that it is production-level ready and ready to be implemented by the masses.

Built With

• facexzoo
• python
• typescript