AI-Powered Innovations at the US Open: A Deep Dive
Project 1: The Content Engine
The US Open tennis tournament is one of the most prestigious sporting events in the world, attracting millions of fans every year. To provide an engaging digital experience for fans, IBM Consulting has been collaborating with the United States Tennis Association (USTA) for over three decades. This year, the teams have come up with two groundbreaking projects that leverage IBM’s versatile family of enterprise-ready Granite foundation models, among other models.
The content engine is one such project, which produces three main outputs: bullet-point descriptive texts before and after every singles match, spoken commentary and subtitles for match highlights, and multi-paragraph Match Reports that provide descriptive summaries and analysis about completed matches.
The Underlying Data
The system draws from a wide range of data points, including world rankings going into the tournament, ongoing match play, and likelihood to win predictions for each singles match. The generative AI system creates pre-match bullet points, which provide insights based on rankings, head-to-head results, and player biographies, giving fans context for the match ahead.
The Generative AI System
The pre-match bullets are generated by a few-shot technique, where the Granite 13b chat model is given examples to follow and deliver similar output. When a match finishes, the system generates text descriptions of what happened, drawn from stats such as aces, break points won, double faults, winners, and shot speed. These descriptions are then transformed into natural language bullet points by generative AI models, including IBM Granite, which are hosted on the IBM Watsonx AI and data platform.
The Benefit
The content engine has revolutionized the way the USTA’s editorial team creates match reports. Before the content engine existed, editors had to spend hours watching replays and interpreting scorelines and stats before they could begin writing longer stories. With the Match Reports, they can now start writing immediately, and for the first time ever, the USTA editorial team will be able to publish a match report for every men’s and women’s singles match this year.
Increased Development Speed and Improved Collaboration with Watsonx Code Assistant
Watsonx Code Assistant provides enterprise-grade code generation, providing snippets and functions to speed application modernization, automation, and scaling. Trained on Granite foundation models, the assistant provides AI-generated recommendations based on existing source code and responds to natural language requests.
Watsonx Code Assistant helped accelerate development of substantial parts of the content engine. Using a code plug-in within their integrated development environment, developers could chat with the assistant through a sidebar panel, asking questions such as how to randomly select text from an array and receiving a recommended code snippet.
Project 2: Audio Commentary
Introduced last year, AI-generated audio commentary provides automated voiceovers and subtitles for every singles match highlight reel shown on the US Open website and app. This feature uses a combination of models, including Granite 13b chat models, to create complex tennis language in support of generated commentary.
Enhancing Personality and Color of Synthetic Speech
This year, the goal was to make the audio commentary more natural and human. The team experimented with two variables: top k, a parameter that controls the number of possible answers the model should consider, and temperature sampling, used to adjust the probability distribution of possible answers. These levers help ensure that the model generates a more human variety of phrases rather than the most probable and repetitive ones.
Testing and Human Review
The teams reviewed and fine-tuned the commentary, striking a balance between artfulness and control. The next step is going from text to speech, where it is essential to make the voices sound convincingly human.
Conclusion
The US Open is a premier sporting event that requires innovative solutions to provide an engaging digital experience for fans. IBM Consulting’s collaboration with the USTA has resulted in two groundbreaking projects that leverage the power of AI to create a more immersive experience. The content engine and audio commentary projects have revolutionized the way the USTA’s editorial team creates match reports and provides commentary, respectively.
FAQs
Q: What is the content engine?
A: The content engine is a system that produces three main outputs: bullet-point descriptive texts before and after every singles match, spoken commentary and subtitles for match highlights, and multi-paragraph Match Reports that provide descriptive summaries and analysis about completed matches.
Q: What is Watsonx Code Assistant?
A: Watsonx Code Assistant provides enterprise-grade code generation, providing snippets and functions to speed application modernization, automation, and scaling. Trained on Granite foundation models, the assistant provides AI-generated recommendations based on existing source code and responds to natural language requests.
Q: What is the goal of the audio commentary project?
A: The goal is to create automated voiceovers and subtitles for every singles match highlight reel shown on the US Open website and app, making the commentary more natural and human.
