Personalized Audio Tagging via Speech Analytics and Feedback

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users are reluctant to add tags to audio conversations due to the effort required and existing automated systems provide non-personalized, inaccurate tag recommendations that do not utilize the audio content effectively.

Innovation Solution

A system that analyzes audio conversations using speech analytics to identify topics and generate personalized tag recommendations based on the conversation content, user history, and context, simplifying the tagging process and increasing the accuracy of tags.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If users manually create tags for audio conversations, then tag accuracy and relevance are improved, but user effort and time consumption increase

Engineering Contradiction:

Improvetag accuracyVSAvoiduser time consumption

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary tag generation by analyzing audio content and generating candidate tags before user interaction. This preliminary action provides users with pre-processed tag options, reducing the time and effort required for manual tagging while maintaining accuracy through user selection or modification of the generated tags.

Inventive Principle:

Principle #10Preliminary action

2Ease of operation

If automated systems generate tags without user input, then user effort is reduced, but tag personalization and accuracy decrease

Engineering Contradiction:

Improveuser effortVSAvoidtag accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The system incorporates user feedback loops where users can review, select, reject, or modify automatically generated tag recommendations. This feedback mechanism allows the system to learn from user preferences and improve future tag generation, achieving both ease of operation through automation and accuracy through user input.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system acts as an intermediary between fully automated tag generation and completely manual tagging. It generates candidate tags based on audio analysis and user profile data, then presents these as recommendations for user review and selection, combining the benefits of both automated and manual approaches.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If tag recommendations are based only on user history, then personalization is improved, but relevance to current audio content decreases

Engineering Contradiction:

ImprovepersonalizationVSAvoidcontent relevance

Core Design Contradiction:

Adaptability or versatilityVSLoss of information

Solution Approach 1:

The system merges multiple information sources including user history, current audio content analysis, conversation context, and topic detection results to generate comprehensive tag recommendations. This combination ensures both personalization through user history and relevance through real-time audio content analysis.

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS9154629B2System and method for generating personalized tag recommendations for tagging audio content

Publication Date: 2015.10.06 ARLINGTON TECHNOLOGIES LLC

AI summary

Systems, methods, and computer-readable storage media for generating personalized tag recommendations using speech analytics. The system first analyzes an audio stream to identify topics in the audio stream. Next, the system identifies tags related to the topics to yield identified tags. Based on the identified tags, the system then generates a tag recommendation for tagging the audio stream. The system can also send the tag recommendation to a device associated with a user for presentation to the user.