Speech-Synchronized Web Session for Multimodal Agent Assistance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multimodal interactions between customers and enterprise agents often become disjointed, leading to poor customer experience and potential loss of business, as human agents cannot view caller activities across multiple interaction channels during self-serve sessions.
Innovation Solution
A computer-implemented method and apparatus that augment voice interactions with speech-synchronized web sessions, allowing human agents to co-browse and provide assistance by capturing contextual information and displaying options to interact with human agents during ongoing sessions, enabling seamless assistance across multiple channels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If self-serve multimodal interaction is provided across multiple channels, then automation and efficiency are improved, but the ability to provide effective assistance deteriorates when issues arise
Solution Approach 1:
The patent introduces a context broker as an intermediary component that bridges the automated self-serve system and human agents. The context broker captures, stores, and manages contextual information from multiple interaction channels (voice, web, mobile), making it available to human agents when assistance is needed. This intermediary ensures seamless transition from automated to human-assisted service while preserving the benefits of both approaches.
Solution Approach 2:
The system performs preliminary actions by capturing and storing contextual information about the caller's interaction journey before human agents become involved. The context broker proactively gathers data from IVR, web sessions, mobile applications, and other channels, preparing the information in advance so that when a human agent is needed, they immediately have access to complete context without requiring the caller to repeat information.
2Adaptability or versatility
If multiple interaction channels are used simultaneously, then customer service capability is improved, but communication coherence deteriorates
Solution Approach 1:
The patent merges information from multiple interaction channels into a unified contextual representation. The context broker consolidates data from voice calls, web browsing sessions, mobile app interactions, and chat communications into a single coherent context profile for each caller. This unified context is then made available across all channels, ensuring that human agents and automated systems have access to the complete interaction history regardless of which channel the customer is using.
Solution Approach 2:
The context broker serves multiple functions across different interaction channels simultaneously. It captures context from various channels, stores it centrally, retrieves it when needed, and provides it to both automated systems and human agents. This universal component ensures that communication coherence is maintained across the entire multimodal interaction ecosystem, allowing the same contextual information to be accessed consistently regardless of the channel or user.
3Reliability
If human agents are connected to assist callers, then assistance quality is improved, but agent awareness of caller activity deteriorates
Solution Approach 1:
The system implements feedback mechanisms where the context broker continuously provides human agents with real-time information about the caller's current activity across different channels. When a caller is browsing a web page, interacting with a mobile application, or navigating through IVR options, this activity information is captured and fed back to the connected human agent through the unified context interface. This enables agents to be fully aware of what the caller is doing and provide contextually relevant assistance.
Data Source
AI summary
A method and apparatus for providing multimodal interaction assistance to customers seeking assistance from agents of an enterprise is disclosed. The method includes augmenting an ongoing voice interaction between a caller and an automated agent with a speech synchronized web session. A session identifier and contextual information in relation to the speech synchronized web session are stored in a database. A display of an option to interact with a human agent is caused during the ongoing speech synchronized web session. In response to a selection of the option by the caller, a co-browsing of the speech synchronized web session by the caller and the human agent is facilitated. The co-browsing of the speech synchronized web session and the contextual information stored in relation to the speech synchronized web session enable the human agent to provide assistance to the caller.