| CPC H04L 65/1086 (2013.01) [H04L 65/403 (2013.01)] | 20 Claims |

|
1. A method comprising:
establishing, by a computing system, a conference call between a plurality of devices, the conference call comprising at least a first and a second breakout session that are being concurrently held at the same time, from a main conference session, wherein a subset of the plurality of devices are automatically moved from the main conference session to each first and second breakout sessions;
generating, by the computing system, a moderator's schedule based at least in part on analyzing keywords from voice samples obtained from the plurality of devices connected to each first and second respective breakout sessions that are being concurrently held at the same time;
generating, by the computing system, a real-time graph that reflects real-time performance in each of the first and second respective breakout sessions, wherein the performance is determined by the computing system based on adherence to an assigned topic for each first and second respective breakout sessions;
determining, by the computing system, that a downward or negative movement in the real-time graph exceeds an acceleration threshold; and
modifying, by the computing system without user intervention, the generated moderator's schedule in response to determining the downward or negative movement in the real-time graph related to the second breakout session, wherein the modified moderator's schedule prioritizes the second breakout session over the first breakout session for a moderator device.
|