Conference call management system
View Patent ↗Systems and methods are described for improving connection to and management of conference calls. An inference/rules engine and human backed speech interpretation subsystem enable the system to efficiently use knowledge about the call to identify the caller and determine to which conference he or she should be connected. During the conference, a virtual assistant is available to interpret and execute system commands spoken by a caller, such as in a side conference accessed via DTMF tones.
1. A system, comprising a processor and a memory in communication with the processor, the memory storing programming instructions executable by the processor to:
enter a virtual assistant mode responsive to receipt of DTMF input from the line of a participant in a conference call;
thereafter mute audio from the line of the participant to the conference call and reduce a volume level of audio from the conference call to the line of the participant;
play spoken input from the participant in a manner perceivable by a first intent analyst in the virtual assistant mode;
accept input from the first intent analyst that indicates the first intent analyst's interpretation of a system request in the spoken input;
provide feedback concerning receipt of the system request to the participant; and
automatically execute the system request responsively to the accepting function;
wherein the programming instructions are further executable by the processor to:
substantially simultaneously with the playing to the first intent analyst, play the spoken input in a manner perceivable by a second intent analyst;
accept input from the second intent analyst that indicates the second intent analyst's interpretation of the system request;
the executing operation is a function of the input accepted from both the first and the second intent analysts;
wherein the programming instructions are further executable by the processor, if the first intent analyst's interpretation and the second intent analyst's interpretation are different, to:
play the spoken input in a manner perceivable by a third intent analyst,
accept input from the third intent analyst that indicates the third intent analyst's interpretation of the system request; and
use the input from the third intent analyst as a tiebreaker between the input from the first intent analyst and the second intent analyst.
2. The system of claim 1 , wherein the system request is in the set of requests consisting of:
muting the line of the of the participant to the conference call;
beginning to transcribe speech in the conference call;
suspending transcription of speech in the conference call;
listing participants in the conference call; and
initiating an outgoing call to attempt to join another person to the conference call.
3. The system of claim 1 , wherein the feedback concerning receipt of the system request is audio feedback.
4. The system of claim 1 , wherein the feedback is prerecorded audio.