How does it work?

The race has two challenges to compete in. You create a racer application that implements REST API and AI we test it out on our racetrack, using our REST API application.

To begin you need to get a client ID and client secret from here and use them to register your racer in the participate page.

Challenge 0: Check name

The first challenge is to verify the display name matches the audio. Whenever you turn on the switch indicating that you are entering the race, we will send aGET request to your /voices/nameresource that should have an audio/mp3 file. Our application will then test to see if the display name matches the mp3 file audio. You will earn a point if the display name matches the mp3 audio.

Challenge 1: Voice Assistant

If the switch is still on, we will send a POST request to your /voices/question/formula-one resource containing an mp3 audio file with a question.

Your racer application is expected to respond with the correct answer as an mp3 audio.

A new question will be sent every 5 minutes. If your racer answers correctly, you will earn 3 points.

Possible question formats include:

• What position did Marc Gené finish in round 17 in 2000?
• Who finished in position 18 in round 5 of 2018?
• Who constructed the car that finished in position 3 in round 1 of 2015?

Challenge 2: Bias Detection

After your racer has answered enough questions correctly in Voice Assistant, we will send a POST request to your/bias-detection/article containing 5 articles. Your racer must determine which article is biased.

If you select the right article, you will earn either 10, 15, or 20 points depending on your response.