Conversation Sentiment Scoring via Transcript Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital communication platforms lack the ability to provide sentiment analysis and analytics data during or after remote communication sessions, making it difficult for sales teams to understand customer sentiment and improve their sales strategies.
Innovation Solution
A system that connects to a communication session, extracts transcripts, identifies relevant utterances, determines word and utterance sentiment scores, and calculates an overall conversation sentiment score, presenting this data to client devices for analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If digital communication platforms record and store conversation transcripts, then the quantity of communication data is increased, but the ability to analyze and extract meaningful insights (sentiment analysis) is lost
Solution Approach 1:
The system extracts sentiment information from conversation transcripts by identifying specific sentiment-bearing words and phrases. The sentiment analysis module separates and extracts only the relevant sentiment data from the large volume of communication data, transforming unstructured text into structured sentiment scores that can be analyzed independently.
Solution Approach 2:
The patent introduces a sentiment analysis module as an intermediary between the communication platform and the user. This module processes the raw conversation data and transforms it into meaningful sentiment metrics, acting as a bridge that converts unstructured communication data into actionable insights without requiring users to manually analyze the entire transcript.
2Measurement precision
If sentiment analysis is performed on all utterances in a communication session, then the precision of sentiment measurement is improved, but the time and computational resources required are increased
Solution Approach 1:
The sentiment analysis process is segmented into multiple stages: first identifying sentiment-bearing words and phrases, then calculating sentiment scores for individual utterances, and finally aggregating these into overall conversation sentiment. This segmentation allows the system to process only the relevant portions of each utterance rather than analyzing every word equally, reducing computational overhead while maintaining precision.
Solution Approach 2:
The system performs sentiment analysis on key sentiment-bearing portions of utterances rather than uniformly processing entire transcripts. By focusing computational resources on identifying and analyzing only the sentiment-critical words and phrases within each utterance, the system achieves accurate sentiment measurement without the time cost of processing every single word in the conversation.
3Adaptability or versatility
If real-time sentiment analysis is provided during communication sessions, then the usefulness of the system for improving sales strategies is enhanced, but the complexity of the system increases
Solution Approach 1:
The sentiment analysis module is designed to work with multiple types of communication data (audio transcripts, chat messages, video captions) and can be applied across different communication scenarios (sales calls, customer service interactions, team meetings). This universal design allows the same core system to serve multiple sales strategy optimization needs without requiring separate specialized systems for each use case.
Solution Approach 2:
The system provides real-time sentiment feedback to sales representatives during communication sessions, allowing them to adjust their approach based on detected customer sentiment. This feedback loop integrates seamlessly into the existing communication workflow, adding analytical capability without requiring a complete restructuring of the sales process or communication platform architecture.
Data Source
AI summary
Sentiment scores are presented within a communication session. In one embodiment, a system extracts, from a transcript, utterances including one or more sentences spoken by the participants. The system identifies a subset of the utterances spoken by a subset of the participants. For each utterance, the system determines a word sentiment score for each word in the utterance, and determines an utterance sentiment score based on the word sentiment scores. The system determines an overall sentiment score for a conversation based on the utterance sentiment scores. The system transmits, to one or more client devices, the overall sentiment score for the conversation.