Audio Intent Summarization for Low-Resource Communication Requests

Solution Overview

Problem

Existing communication systems consume significant network resources, including power, memory, and processing resources, due to lengthy and large data exchanges between devices.

Innovation Solution

A system and method that generates information requests based on audio data exchanged between user and workspace devices, using machine learning algorithms to transcribe and summarize the data, determine intent, and provide action item suggestions to optimize communication operations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional data exchange communication operations are performed between devices, then complete information transfer is achieved, but network resources including power, memory, and processing resources are significantly consumed

Engineering Contradiction:

Improveinformation transfer completenessVSAvoidnetwork resource consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system extracts only the essential intent and key information from audio data using machine learning algorithms, rather than transmitting or processing the complete audio data. This extraction approach maintains information transfer completeness while significantly reducing network resource consumption by focusing only on the most relevant extracted features.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs preliminary transcription and intent determination on audio data at the source device before transmission. By pre-processing the audio data to extract intent and summarize key points, the system reduces the amount of data that needs to be transmitted over the network, thereby reducing network resource consumption while maintaining complete information transfer.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If lengthy communication operations lasting multiple minutes are performed, then complete data exchange is achieved, but network resources are consumed over extended periods

Engineering Contradiction:

Improvedata exchange completenessVSAvoidcommunication operation duration

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary transcription and intent extraction on audio data at the source device before transmission. By pre-processing the audio data to extract intent and summarize key points, the system reduces the amount of data that needs to be transmitted over the network, thereby reducing network resource consumption while maintaining complete information transfer.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Instead of transmitting the original audio data, the system creates and transmits a simplified copy in the form of extracted intent and text summary. This copy contains the essential information needed for communication while being significantly smaller and faster to transmit, reducing both communication duration and resource consumption.

Inventive Principle:

Principle #26Copying

3Reliability

If large data exchanges involving multiple information packets are performed, then complete information transfer is achieved, but network resources including memory and processing resources are consumed

Engineering Contradiction:

Improveinformation transfer completenessVSAvoiddata exchange volume

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The system extracts only the essential intent and key information from audio data using machine learning algorithms, rather than transmitting or processing the complete audio data. This extraction approach maintains information transfer completeness while significantly reducing network resource consumption by focusing only on the most relevant extracted features.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of transmitting the original audio data, the system creates and transmits a simplified copy in the form of extracted intent and text summary. This copy contains the essential information needed for communication while being significantly smaller and faster to transmit, reducing both communication duration and resource consumption.

Inventive Principle:

Principle #26Copying

Data Source

PatentUS20250390681A1System and method to generate information requests based on audio data

Publication Date: 2025.12.25 BANK OF AMERICA CORP

AI summary

A system comprises a memory communicatively coupled to at least one processor. The processor is configured to obtain audio data from a user device configured to perform one or more communication operations with a workspace device. In response to receiving the audio data, the processor is configured to execute the machine learning algorithm to transcribe the audio data into text data and summarize the text data into a request summary. Further, the processor is configured to determine a target operation based on the request summary. The target operation is a determined intent to perform a communication operation. The processor is configured to determine whether the communication operation at least partially matches the authorized communication operations and present the request summary as a reset point to train the one or more machine learning models in response to determining that the communication operation at least partially matches the authorized communication operations.