Dynamically generated topic segments for a communication session

1. A method, comprising:

receiving a list of topics;

receiving a conversation transcript that includes one or more transcriptions of utterances transmitted between a plurality of participant user accounts during a communication session comprising an online virtual meeting, the transcript comprising at least one timestamp corresponding to a time within a time duration of the online virtual meeting at which a respective utterance occurred;

for each topic in the list of topics, segmenting the one or more utterances into one or more topic segments based on the topic, wherein the segmenting includes determining an utterance boundary based on a lexical score that is an inner product of two vectors associated with an adjacent pair of text blocks, wherein a vector contains a number of times a lexical item occurs within a corresponding text block;

for each of the topic segments:

classifying whether the respective topic segment is related to the topic;

determining that the respective topic segment is related to the topic; and

identifying a respective start time and a respective end time of each respective topic segment; and

transmitting, to one or more client devices, a list of the topic segments, the list of topic segments including a start and an end time for at least one of the topic segments.

2. The method of claim 1 , further comprising:

if the segment is related to the topic, generating a title for the topic segment based on the topic.

3. The method of claim 1 , wherein the list of topics is received from a client device associated with an authorized user.

4. The method of claim 1 , wherein the segmenting is performed via one or more text tiling techniques.

5. The method of claim 1 , wherein the segmenting comprises:

shifting a window over the utterances in the transcript one word at a time with a pre-specified window size to generate two blocks of utterances per each shift of the window;

at each shift of the window, comparing the two blocks of the utterances to determine whether the blocks are semantically similar; and

defining a boundary between two topic segments when two blocks of utterances are semantically different.

6. The method of claim 1 , wherein at least a subset of the topic segments overlap with one or more of the other topic segments.

7. The method of claim 1 , wherein the topic segments comprise a span of the transcript comprising one or more lines or utterances.

8. The method of claim 1 , wherein classifying whether the topic segment is related to the topic is performed via one or more zero-shot text classification techniques.

9. The method of claim 1 , wherein classifying whether the topic segment is related to the topic comprises meeting or exceeding a relatedness threshold.

10. The method of claim 1 , wherein the segmentation is performed via linear segmentation for each topic.

11. The method of claim 1 , wherein classifying whether the topic segment is related to the topic is based on one or more language models.

12. The method of claim 1 , wherein classifying whether the topic segment is related to the topic is based on one or more keywords.

13. The method of claim 1 , further comprising:

transmitting, to one or more client devices, a topic summary for a topic, the topic summary comprising one or more utterances from topic segments related to the topic.

14. The method of claim 1 , further comprising:

transmitting, to one or more client devices, one or more utterance results based on a search for the topic within the communication session.

15. The method of claim 1 , further comprising:

transmitting, to one or more client devices, analytics data related to one or more topics within the communication session.

16. A communication system comprising:

a processor configured to:

receive a list of topics;

receive a conversation transcript that includes one or more transcriptions of utterances transmitted between a plurality of participant user accounts during a communication session comprising an online virtual meeting, the transcript comprising at least one timestamp corresponding to a time within a time duration of the online virtual meeting at which a respective utterance occurred;

for each topic in the list of topics, segment the one or more utterances into one or more topic segments based on the topic, wherein the processor is configured to determine an utterance boundary based on a lexical score that is an inner product of two vectors associated with an adjacent pair of text blocks, wherein a vector contains a number of times a lexical item occurs within a corresponding text block;

for each of the topic segments, the processor further configured to:

classify whether the respective topic segment is related to the topic;

determine that the respective topic segment is related to the topic; and

identify a respective start time and a respective end time of each respective topic segment; and

transmit, to one or more client devices, a list of the topic segments, the list of topic segments including a start and an end time for at least one of the topic segments.

17. The communication system of claim 16 , wherein the processor is further configured to segment the one or more utterances using one or more topic matching techniques.

18. A non-transitory computer-readable medium containing instructions that when executed by a processor, cause the processor to perform operations including:

receiving a conversation transcript that includes one or more transcriptions of utterances transmitted between a plurality of participant user accounts during a communication session comprising an online virtual meeting, the transcript comprising at least one timestamp corresponding to a time within a time duration of the online virtual meeting at which a respective utterance occurred;

for each topic in the list of topics, segmenting the one or more utterances into one or more topic segments based on the topic, wherein the segmenting includes determining an utterance boundary based on a lexical score that is an inner product of two vectors associated with an adjacent pair of text blocks, wherein a vector contains a number of times a lexical item occurs within a corresponding text block;

for each of the topic segments:

classifying whether the respective topic segment is related to the topic;

determining that the respective topic segment is related to the topic; and

identifying a respective start time and a respective end time of each respective topic segment; and

transmitting, to one or more client devices, a list of the topic segments, the list of topic segments including a start and an end time for at least one of the topic segments.

19. The non-transitory computer-readable medium of claim 18 , wherein classifying whether the topic segment is related to the topic is performed via one or more zero-shot text classification techniques.

20. The non-transitory computer-readable medium of claim 18 , wherein classifying whether the topic segment is related to the topic comprises meeting or exceeding a relatedness threshold.