| CPC G06F 16/9535 (2019.01) [G10L 15/08 (2013.01); G10L 2015/088 (2013.01)] | 20 Claims |
|
1. A method, comprising:
receiving a list of content recommendation actions specifically adjusted for a particular type of communication session, each content recommendation action being associated with one or more trigger phrases comprising conditions for performance of the content recommendation action during the particular communication session, each trigger phrase being associated with a party the trigger phrase is to be uttered by, wherein the content recommendation actions and corresponding trigger phrases are generated based on actions and trigger phrases performed by other users of a communication platform;
connecting to the communication session with a plurality of participants;
receiving a plurality of utterances associated with the participants in real time during the communication session, wherein each utterance is associated with a speaker;
for each utterance, determining, by one or more pre-trained language learning models, whether a prediction of relatedness is present between the utterance and one or more trigger phrases associated with a content recommendation action from the list of content recommendation actions;
for each utterance, determining whether the speaker associated with the utterance is the party associated with the trigger phrase and the content recommendation action; and
upon determining that a prediction of relatedness is present and that the speaker is the party, performing the associated content recommendation action by transmitting, to one or more client devices, one or more pieces of content for display on the one or more client devices during the communication session, wherein the one or more pieces of content includes the corresponding trigger phrase.
|