Call Coordination System Using Voice Analysis for Representative Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current call coordination technologies fail to effectively utilize advanced speech recognition, natural language processing, and machine learning to optimize the pairing of sales representatives and customers based on conversation characteristics, leading to suboptimal call outcomes.
Innovation Solution
A call-coordination system that analyzes voice conversations using offline and real-time components to generate features and classifiers, which are then used to route calls to the most suitable representative, maximizing the probability of desired outcomes by leveraging automatic speech recognition, natural language processing, and machine learning algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current call coordination technologies are used (based on representative experience and geographical location), then implementation is simple and quick, but call outcomes are suboptimal and coordination effectiveness is low
Solution Approach 1:
The system performs preliminary analysis of conversation characteristics during early call stages, extracting features and generating classifiers before the call concludes. This allows the system to make informed routing decisions based on analyzed conversation data rather than relying solely on pre-call representative experience or post-call analysis, thereby improving call outcomes while managing complexity through phased processing.
Solution Approach 2:
The patent introduces an intermediary call coordination system that acts as a mediator between customers and representatives. This intermediary analyzes conversation characteristics in real-time and makes intelligent routing decisions, bridging the gap between simple geographic routing and complex outcome optimization. The intermediary processes conversation data and translates it into actionable routing recommendations, improving effectiveness without requiring direct complex integration into existing representative workflows.
2Measurement precision
If advanced speech recognition and natural language processing are implemented to analyze conversation characteristics, then coordination accuracy improves, but computational resources and processing time increase
Solution Approach 1:
The system applies partial action by focusing computational resources on analyzing only the most relevant conversation characteristics rather than processing every aspect of the conversation equally. The feature extraction and classifier generation target specific outcome-influencing elements identified through offline analysis, applying advanced NLP and speech recognition selectively to key segments and features that most impact call outcomes, thereby reducing overall computational burden while maintaining high accuracy.
Solution Approach 2:
The system performs preliminary offline analysis to pre-compute features and generate classifiers before real-time call processing. This preliminary action creates reusable models and feature templates that can be applied during actual calls without requiring full re-analysis, significantly reducing real-time computational resource consumption while maintaining high measurement precision through pre-validated analysis frameworks.
3Productivity
If real-time call data analysis is performed to generate features and classifiers, then call routing optimization improves, but processing time and system response time increase
Solution Approach 1:
The system performs preliminary offline analysis to pre-compute conversation features and generate classifiers before real-time call processing. This allows the system to have pre-ready models and feature templates that can be quickly applied during actual calls. The offline component creates a library of analyzed patterns and routing recommendations that the real-time component can rapidly retrieve and apply, minimizing real-time processing delays while maintaining routing optimization quality.
Solution Approach 2:
The patent segments the call coordination system into distinct offline and real-time components with separate responsibilities. The offline component handles computationally intensive feature extraction and classifier generation during non-peak periods, while the real-time component focuses on rapid inference and routing decisions. This segmentation allows each component to be optimized independently, with the offline portion performing thorough analysis and the real-time portion delivering fast responses, thereby improving overall routing efficiency without compromising processing speed.
Data Source
AI summary
The disclosure is directed to analyzing voice conversations between participants of conversations and coordinating calls between participants, e.g., in order to influence an outcome of the voice conversation. For example, sales calls can be coordinated between specific sales representatives (“representatives”) and customers by routing a sales call from a customer to a specific sales representative, based on their voices and the content of the conversation, with the goal of positively influencing the outcome of the sales call. A mapping between sales representatives and customers that is set to maximize the probability for certain outcomes is generated. This mapping (or pairing) may be fed into either an automatic or manual coordination system that connects or bridges sales representatives with customers. The mapping may be generated either based on historic data or early-call conversation analysis, in both inbound and outbound calls.