Abstract
This study explores the use of generative artificial intelligence (Generative AI) in virtual sportscasting for grassroots baseball in Taiwan. It focuses on how AI-generated tone and real-time game data influence audience immersion, emotional engagement, and trust. A mixed-methods approach was used, combining surveys and Reflexive Thematic Analysis (RTA), with participants being parents from a youth baseball team in Hsinchu. They viewed three types of commentary: neutral, supportive, and supportive with data, and rated perceived excitement and style. Results show strong audience acceptance of AI sportscasters, especially those using advanced statistics, which improved credibility and clarity. Three key themes emerged: emotional-tone mismatch, data trust and comprehension, and voice authenticity. The study highlights the need to balance tone with data flow to avoid disrupting pacing and to enhance viewer experience. The findings suggest that effective virtual sportscasting depends on the synergy between tone and data. Emotional tone or raw data alone is not enough— adaptive modulation of both is needed for immersive engagement. The study proposes three modules: (1) expandable tone settings, (2) key-event data filtering, and (3) audience-type adjustment. This research supports Taiwan’s “National Baseball Initiative” and recommends joint development of local AI broadcasting platforms to increase visibility and engagement in community sports.
Keywords
Generative AI; Sportscasting; Grassroots baseball
DOI
https://doi.org/10.21606/iasdr.2025.457
Citation
Chen, Y.C., University, T.H., Mah, K.H.,and Tseng, Y.(2025) Generative AI-Based Virtual Sportscaster for Baseball: the Influence of In-Game Data and Broadcasting Style, in Chang, C.-Y., and Hsu, Y. (eds.), IASDR 2025: Design Next, 02-05 December, Taiwan. https://doi.org/10.21606/iasdr.2025.457
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
Conference Track
Track 4 - Human-Centered AI
Generative AI-Based Virtual Sportscaster for Baseball: the Influence of In-Game Data and Broadcasting Style
This study explores the use of generative artificial intelligence (Generative AI) in virtual sportscasting for grassroots baseball in Taiwan. It focuses on how AI-generated tone and real-time game data influence audience immersion, emotional engagement, and trust. A mixed-methods approach was used, combining surveys and Reflexive Thematic Analysis (RTA), with participants being parents from a youth baseball team in Hsinchu. They viewed three types of commentary: neutral, supportive, and supportive with data, and rated perceived excitement and style. Results show strong audience acceptance of AI sportscasters, especially those using advanced statistics, which improved credibility and clarity. Three key themes emerged: emotional-tone mismatch, data trust and comprehension, and voice authenticity. The study highlights the need to balance tone with data flow to avoid disrupting pacing and to enhance viewer experience. The findings suggest that effective virtual sportscasting depends on the synergy between tone and data. Emotional tone or raw data alone is not enough— adaptive modulation of both is needed for immersive engagement. The study proposes three modules: (1) expandable tone settings, (2) key-event data filtering, and (3) audience-type adjustment. This research supports Taiwan’s “National Baseball Initiative” and recommends joint development of local AI broadcasting platforms to increase visibility and engagement in community sports.