SAP has unveiled an 'AI-based Voice Intelligence' application developed in collaboration with the professional esports team Team Liquid. This application transforms unstructured in-game voice communication data into structured insights and sorts them chronologically. The explanation is that it converts esports communication data, which has previously accumulated only in audio form and was difficult to analyze systematically, into a usable format.

SAP announced this on the 23rd, stating that coaches can now analyze manual reviews that previously took hours in just a few minutes.

In competitive esports like League of Legends, team communication is constant, but the multitrack voice data generated by professional teams during training and competitions is siloed into audio only, making large-scale analysis virtually impossible.

The application automates end-to-end processing of voice communication data.

Once recordings of matches or training are secured, multi-channel audio is collected and individual player voice streams are separated.

Then, segments are divided based on voice activity detection (VAD), converted into text using the Gemini 2.5 Pro API, and then the speaker is identified and emotions are analyzed to generate structured data.

The coaching staff can identify the communication leader, points of disagreement, and emotional changes at critical moments.

SAP stated that it can be used to detect a player's frustration or emotional tilt early on, and to evaluate the chemistry of a new roster.

Jessie Hart (Dr. Jesse Hart, Senior Director of Sports Science and Analytics at Team Liquid, stated, “This is the first time we have been able to systematically analyze team communication under the pressure of a real game environment,” adding, “We can analyze game situations and results in conjunction with communication, emotions, and decision-making processes.”

The application runs on the SAP Business Technology Platform (SAP BTP).

SAP HANA Cloud stores structured data and supports sentiment analysis within the database, and AI models are managed through SAP AI Core.

The explanation is that it reduces data movement and latency by managing speaker identification, text conversion, sentiment analysis results, and metadata in a single system.

Dr. Benjamin Blau, SAP’s Chief Process and Information Officer (CPIO), stated, “To date, we have processed thousands of audio tracks from Team Liquid and provided searchable, sentiment-rated insights within minutes.”