This week I found myself at Wimbledon and Centre Court watching the likes of Novak Djokovic, Venus Williams and Andrey Rublev. But there's much more than the tennis to see...
I came across a stand by IBM who were offering an AI data informed commentary experience. Artifical Intelligence (AI) Commentary is a new feature for 2023 that adds spoken AI-generated narrative to the video highlight reels on Wimbledon.com. IBM worked with The All England Club to leverage foundation models from watsonx, IBM's enterprise AI and data platform, to train the AI in the unique language of tennis - impressive eh? Both male and female generative commentary is provided alongside captions. Generative AI built on these foundation models was applied to produce narration with varied sentence structure and vocabulary to make the clips informative and engaging. AI is moving quicker than Novak Djokovic or Carlos Alcaraz, and that's saying something!
However, I don't think Andrew Castle, Martina Navratilova or Tim Henman will be losing their jobs anytime soon as Andy Murray had his say in a recent interview : "I'd prefer my commentators to be doing it live but yeah I'll be interested to see what the AI commentary is like."
Anyway, I was invited to have a go myself but I was a little unprepared. I could choose from 12 clips of Wimbledon matches from last year to commentate on and while the clip was playing on one screen, on the left hand side I was provided with a wealth of tennis data to choose from. I learned that the data is generated from cameras they have tracking every move of the players, every swing, every step and every hit (with the assistance of Hawk-eye). Their AI uses ball tracking and player data to identify types of shots being used too! I then had 2 minutes to provide come commentary with the data and statistics provided to spruce it up. I froze completely before saying some generic tennis words (information overload!). I'd be better at football commentary honestly, I promise!
IBM also uses statistical modelling to map out chance of progression through the tournament and other tennis-related stats (My only complaint was that this was data overload!) This feature is called AI Draw Analysis, which utilises the IBM Power Index and likelihood to Win predictions to assess each player’s potential path to the final. As soon as any tournament draw is released, players and fans alike intuitively assess each player’s luck and path through the field: do they have a “good draw” or a “bad draw”? This is marked by 'Favourable' and 'Difficult' (Politely described and in keeping with the Wimbledon brand rather than 'Easy Win' and ' Not a chance in hell'). This year, IBM AI Draw Analysis helps them make more data-informed predictions by providing a statistical factor (a draw ranking) for each player in the singles draw.
This illustrates how quickly AI has driven and changed the fan's experience of tennis in a data-driven way and it is hard not to be impressed by the sheer amount of data produced in such a short space of time. Now if IBM could just sort out the queues...
Second serve: Are you a tennis fan?
I love tennis!
It's something to watch :/
Anything but tennis!