en.Wedoany.com Reported - SAP announced on the 23rd the official launch of an AI-based voice intelligence application developed in collaboration with Team Liquid. The application converts unstructured in-game voice communication data into structured game insights, presented in chronological order. This enables coaching teams to systematically complete analysis in just a few minutes, a process that previously required hours of manual review.
In competitive esports matches such as *League of Legends*, real-time team communication is high-frequency and high-intensity, with a decisive impact on match performance. However, hours of multi-track voice data were previously difficult to analyze at scale and often existed only as isolated audio files. The application automates the end-to-end processing of voice communication data. After obtaining match or practice recordings, it automatically collects multi-channel audio, separates each player's voice stream, converts the communication into text, identifies speakers, performs sentiment analysis, and ultimately generates structured data usable by the team.
With this application, Team Liquid can obtain objective and consistent communication data for each match. This allows the coaching team to understand who dominated the communication, where disagreements occurred, and how emotional states changed during critical moments. The application also supports early detection of player frustration or emotional dysregulation, helping them make clearer calls and decisions under pressure, and can be used to evaluate the synergy of new lineups over time. Dr. Jesse Hart, Senior Director of Sports Science and Analytics at Team Liquid, stated that this is the first time team communication can be systematically analyzed under the pressure of a real competitive environment. The application transforms raw voice chat data into valuable insights, enabling the coaching team to correlate match context and outcomes with communication, emotions, and decision-making processes.
The AI application runs on the SAP Business Technology Platform (SAP BTP). SAP HANA Cloud serves as the architectural core, storing all structured data and supporting in-database sentiment analysis. By managing speaker identification information, text conversion, sentiment analysis results, and metadata within a unified management system, the solution minimizes data movement, reduces latency, and ensures traceability. AI models are managed through SAP AI Core. The processing workflow begins with multi-channel audio collection, followed by segmentation based on voice activity detection, AI transcription using the Gemini 2.5 Pro API, and quality verification to generate final transcription data. Dr. Benjamin Blau, Chief Process and Information Officer at SAP, revealed that the application has processed thousands of audio tracks for Team Liquid, providing searchable insights with sentiment scores within minutes. SAP and Team Liquid stated that this collaboration is an example of AI and analytics technology applied in professional esports, demonstrating how unstructured voice data can be transformed into actionable intelligence in high-pressure competitive environments.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
