Multimodal Conversation Analysis for Coaching Effectiveness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in effectively analyzing and leveraging multiparty conversation data to develop employee skills through coaching relationships, especially in geographically distributed workforces.
Innovation Solution
A machine learning system that analyzes acoustic, video, and text data from multiparty conversations to determine conversation analysis indicators, using multimodal and sequential machine learning techniques to synthesize data across modalities and over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If machine learning systems analyze multiparty conversation data to develop employee skills, then coaching effectiveness is improved, but system complexity increases
Solution Approach 1:
The system segments the complex conversation analysis task into distinct processing stages: data collection from multiple modalities (acoustic, video, text), feature extraction for each modality, synthesis of features across modalities, sequential processing over time, and generation of conversation analysis indicators. This segmentation allows the complex system to be managed through modular components that each handle a specific aspect of the analysis.
Solution Approach 2:
The patent implements nested processing where lower-level features from different modalities are synthesized to create higher-level conversation features, which are then sequentially processed to generate final conversation analysis indicators. This nested structure allows complex analysis to be built incrementally from simpler components, with each layer building upon the previous one.
2Adaptability or versatility
If geographically distributed workforces are used, then organizational flexibility is improved, but coaching delivery becomes more difficult
Solution Approach 1:
The system enables self-service coaching by automatically analyzing conversation data and generating conversation analysis indicators without requiring manual intervention. The machine learning system performs the coaching analysis autonomously, processing acoustic, video, and text data to provide insights that coaches can use to develop employee skills, eliminating the need for manual review of extensive conversation recordings.
Solution Approach 2:
The patent replaces manual coaching review processes with automated machine learning systems that process and analyze conversation data. Instead of coaches manually reviewing hours of video and audio recordings, the system uses machine learning algorithms to automatically extract features, synthesize information across modalities, and generate actionable insights, substituting mechanical manual analysis with automated computational processes.
3Measurement precision
If multiple data modalities are analyzed, then conversation analysis precision is improved, but data processing complexity increases
Solution Approach 1:
The system merges multiple data modalities (acoustic, video, and text) by extracting features from each modality separately and then synthesizing these features into a unified representation of conversation characteristics. This merging process allows the system to leverage the complementary strengths of different data types while managing complexity through a structured integration approach.
Solution Approach 2:
The machine learning system is designed with multi-functionality to handle multiple data modalities through a unified framework. The same sequential processing mechanism can analyze features from acoustic, video, and text sources, generating conversation analysis indicators that integrate information from all modalities. This universal approach simplifies the processing of diverse data types by using a common analytical pipeline.
Data Source
AI summary
Technology is provided for causing a computing system to extract conversation features from a multiparty conversation (e.g., between a coach and mentee), apply the conversation features to a machine learning system to generate conversation analysis indicators, and apply a mapping of conversation analysis indicators to actions and inferences to determine actions to take or inferences to make for the multiparty conversation. In various implementations, the actions and inferences can include determining scores for the multiparty conversation such as a score for progress toward a coaching goal, instant scores for various points throughout the conversation, conversation impact score, ownership scores, etc. These scores can be, e.g., surfaced in various user interfaces along with context and benchmark indicators, used to select resources for the coach or mentee, used to update coach/mentee matchings, used to provide real-time alerts to signify how the conversation is going, etc.