Call Center Speech Routing for Multi-Customer Agent Service
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing customer service systems in virtual environments, such as the metaverse, are limited by the inability of a single agent to simultaneously interact with multiple customers due to constraints in voice and video interactions, leading to waiting times and inefficient resource management.
Innovation Solution
A system utilizing text-to-speech (TTS) and automatic speech recognition (ASR) technologies to convert voice communications to text and vice versa, enabling a single agent to communicate with multiple customers through a chat interface, supplemented by a speech generator to simulate natural conversation and manage delays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If an agent uses traditional voice communication to interact with customers, then the communication is natural and seamless, but the agent can only handle one customer at a time
Solution Approach 1:
The patent introduces text-to-speech (TTS) and automatic speech recognition (ASR) technologies as intermediaries between the agent and customers. The agent types text messages, TTS converts them to speech for customer hearing, and ASR converts customer speech to text for the agent to read. This mediator system enables one-to-many communication while preserving natural voice interaction for customers.
Solution Approach 2:
The patent replaces the traditional mechanical voice-to-voice communication system with a hybrid system combining text input, TTS conversion, and ASR recognition. This substitution allows the agent's single text input to be converted into multiple speech outputs simultaneously directed to different customers, breaking the one-to-one communication constraint.
2Productivity
If the system uses TTS to convert agent text to speech for multiple customers, then the agent can serve multiple customers simultaneously, but the system complexity increases
Solution Approach 1:
The patent makes the TTS engine a universal component that serves multiple customers simultaneously. Instead of having separate TTS instances for each customer, a single TTS engine processes the agent's text input and generates speech outputs for multiple customers, reducing redundancy and managing complexity through shared resources.
Solution Approach 2:
The patent merges the communication channels by combining TTS and ASR functionalities into an integrated system architecture. The call center server consolidates multiple communication streams, routing agent text inputs through TTS to multiple customers and aggregating customer speech inputs through ASR to the agent, simplifying the overall system structure.
3Productivity
If the call center server routes multiple customer devices to a single agent device, then customer service capacity increases, but communication management becomes more difficult
Solution Approach 1:
The patent segments the communication management into distinct functional modules: text input processing, TTS conversion, speech routing to multiple customers, ASR recognition, and text output to the agent. This segmentation allows the call center server to handle multiple customer connections through organized, manageable components rather than monolithic complexity.
Solution Approach 2:
The call center server acts as an intermediary that manages the many-to-one communication routing. It receives text from the agent, distributes TTS-generated speech to multiple customers, collects ASR-converted text from customers, and routes it back to the agent. This mediator role simplifies routing management by centralizing control logic.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables a single agent to serve multiple customers seamlessly, enhancing customer experience by reducing wait times and improving interaction efficiency in virtual environments.
Implementation Method 1
a text-to-speech (TTS) engine that converts text entered into the agent device into speech that is transmitted by the call center server to one or more of the plurality of customer devices
Implementation Method 2
automatic speech recognition (ASR) to convert customer voice into text
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer system and method are configured to permit a single agent to simultaneously interact with multiple customers. The computer system includes a call center server configured to communicate with a plurality of customer devices, and an agent device. A text-to-speech (TTS) engine converts text entered into the agent device into speech that is transmitted by the call center server to at least some of the plurality of customer devices. Convert SMS emails and chats messages from agent to customer devices. Pause conversation if detected that agent needs more time. Fill pauses and interruptions of the agent with a voice filler.