Multi-service business platform system having conversation intelligence systems and methods
The disclosure is directed to various ways of improving the functioning of computer systems, information networks, data stores, search engine systems and methods, and other advantages. Among other things, provided herein are methods, systems, components, processes, modules, blocks, circuits, sub-systems, articles, and other elements (collectively referred to in some cases as the “platform” or the “system”) that collectively enable, in one or more datastores (e.g., where each datastore may include one or more databases) and systems. A system and method for providing conversation intelligence services may include pre-processing, transcribing, and post-processing. A conversation recording may be pre-processed generating a conversation record (e.g., conversation object). The pre-processed conversation recording may be transcribed into a transcript. The transcript may be post-processed which may include keyword extraction, topic extraction, feature extraction, event generation, trigger action, and/or search indexing. Conversation information may be presented based on the pre-processing, the transcribing, and the post-processing.
1. A computer-implemented method comprising:
processing a conversation recording to determine speaker identities and to generate a conversation record;
transcribing the conversation recording into a transcript based on the conversation record;
extracting features from the transcript based on data in the transcript and the conversation record, wherein the features are extracted using search indexing based on data in the transcript and the conversation record, wherein the features are extracted using a sentiment model reinforced by a machine learning module that parses transcripts to create training data pairs of features and labels for training and reinforcing the sentiment model; and
in response to the features extracted from the transcript indicating that an occurrence of an event occurred, generating an event notification specifying a type of the event that occurred;
presenting conversation information, including the event notification, in a graphical user interface based on the speaker identities, transcript, and the features.
2. The method of claim 1 , wherein the conversation recording includes media files that each correspond to a different speaker in the conversation.
3. The method of claim 1 , wherein the processing further includes obtaining information relating to speakers identified by the speaker identities, wherein the information includes a name of each speaker, a title of each speaker, and a company of each speaker.
4. The method of claim 1 , wherein the conversation recording is processed to generate a quality review for improving future transcription, wherein the quality review includes accessing information referred to as transcription hints for improving future transcription, and wherein the transcription hints include a list of words likely to appear in the transcript of the conversation.
5. The method of claim 1 , further comprising performing the search indexing that includes indexing the transcript such that the conversation of the transcript is searchable via a search service and keywords from the transcript based on a list of identified keywords.
6. The method of claim 1 , further comprising performing the search indexing that includes indexing the transcript such that the conversation of the transcript is searchable via a search service and classifying topics discussed during the conversation based on features of the transcript.
7. The method of claim 1 , wherein the extracting comprises feature extraction that includes keyword extraction, topic extraction, feature extraction, event generation, and trigger action, and wherein the feature extraction includes extracting a tone and a pace of the conversation.
8. The method of claim 1 , further comprising performing event generation that includes generating an event notification when a feature is detected in the transcript, and wherein the event notification indicates a type of event, speakers involved in the event, and metadata related to the event.
9. The method of claim 1 , further comprising triggering of a workflow.
10. A system including a processor and memory storing instructions which when executed by the processor perform operations comprising:
processing a conversation recording to determine speaker identities and to generate a conversation record;
transcribing the conversation recording into a transcript based on the conversation record;
extracting features from the transcript based on data in the transcript and the conversation record, wherein the features are extracted using search indexing based on data in the transcript and the conversation record, wherein the features are extracted using a sentiment model reinforced by a machine learning module that parses transcripts to create training data pairs of features and labels for training and reinforcing the sentiment model; and
in response to the features extracted from the transcript indicating that an occurrence of an event occurred, generating an event notification specifying a type of the event that occurred;
presenting conversation information, including the event notification, in a graphical user interface based on the speaker identities, transcript, and the features.
11. The system of claim 10 , wherein the operations comprise indexing the transcript such that the conversation of the transcript is searchable via a search service and extracting keywords from the transcript based on a list of identified keywords.
12. The system of claim 10 , wherein the operations comprise indexing the transcript such that the conversation of the transcript is searchable via a search service and classifying topics discussed during the conversation based on features of the transcript.
13. The system of claim 10 , wherein the operations comprise keyword extraction, topic extraction, feature extraction, and event generation.
14. The system of claim 10 , wherein the operations comprise generating an event notification when a keyword is detected in the transcript, and wherein the event notification indicates a type of event, speakers involved in the event, and metadata related to the event.
15. The system of claim 10 , wherein the operations comprise performing a trigger action as part of the post-processing, wherein the trigger action includes triggering a service of a multi-service platform.
16. A non-transitory computer readable storage medium storing instructions which when executed by a processor cause the processor to perform operations comprising:
processing a conversation recording to determine speaker identities and to generate a conversation record;
transcribing the conversation recording into a transcript based on the conversation record;
extracting features from the transcript based on data in the transcript and the conversation record, wherein the features are extracted using search indexing based on data in the transcript and the conversation record, wherein the features are extracted using a sentiment model reinforced by a machine learning module that parses transcripts to create training data pairs of features and labels for training and reinforcing the sentiment model; and
in response to the features extracted from the transcript indicating that an occurrence of an event occurred, generating an event notification specifying a type of the event that occurred;
presenting conversation information, including the event notification, in a graphical user interface based on the speaker identities, transcript, and the features.
17. The computer readable storage medium of claim 16 , wherein the operations further comprise search indexing that includes indexing the transcript such that the conversation of the transcript is searchable via a search service and extracting keywords from the transcript based on a list of identified keywords.
18. The computer readable storage medium of claim 16 , wherein the operations further comprise search indexing that includes indexing the transcript such that the conversation of the transcript is searchable via a search service and classifying topics discussed during the conversation based on features of the transcript.
19. The computer readable storage medium of claim 16 , wherein the operations further comprise event generation that includes generating an event notification when a feature is detected in the transcript, and wherein the event notification indicates a type of event, speakers involved in the event, and metadata related to the event.
20. The computer readable storage medium of claim 16 , wherein the operations further comprise triggering of a task.