CRIC-TALK

Artificial Intelligence (AI) is an umbrella term covering a variety of what are called “smart” technologies. What they all have in common is the ability to learn.

The use of artificial intelligence in sports isn’t a major surprise, given the advancement in technology. The rise in computing power, availability of massive amounts of data, and an increased willingness of stakeholders to leverage such tools are the three principal reasons why the role of artificial intelligence in sports has gained a lot more importance recently. Today every cricket lover has the freedom to analyze every ball and every shot from all possible angles, analyze how the teams fared, new milestones created, how far the ball traveled into the stands, what pace the bowlers bowled and so on – all from the comfort of their home. The impact of technology seems to have completely changed cricket’s game plan.

However, the automated commentary is still a fantasy and the way AI and ML are developing, it could soon be a reality. The development of such a system will change the way of sports broadcasting. Hence, in this work, we have tried to generate a part of the automatic ball-by-ball commentary of a cricket match.

We were studying for our PGP in Data Science, Business Analytics, and Big Data at the Aegis School of Data Science, Telecommunication, and Cyber Security. We were nearing the end of our semester and needed to complete our final Capstone project, which would demonstrate the skills and techniques we had learnt by developing an end-to-end product.

We had previously considered doing something related to finance because we both have a Commerce educational background and a demonstrated experience of working in the banking industry. However, we wanted to try something new and push ourselves beyond our comfort zone. We wanted to produce something about cricket because we are both sports fans, more specifically cricket fans. After researching the work of Artificial Intelligence in the field of Cricket, we discovered a portion of it that is still unexplored, therefore we decided to take on this task and attempt to Generate Automatic Cricket Commentary Using AI.

Our goal was to create CNN models that could recognize and locate the players, as well as identify their identities and the shots they took. And we planned to create a final commentary based on these factors.

In this work, we have tried our hands-on Deep Learning models for the purpose of classification and object detection/location models.