ML Record Generation From Audio Transcripts in Collaboration Workflows

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Manual creation of work unit records from recorded audio/video in collaboration environments is time-consuming and prone to user error, decreasing workflow efficiency.

Innovation Solution

A system utilizing a machine learning model trained on a text corpus to automatically generate records from asynchronously recorded audio and/or video, extracting and structuring content to facilitate the creation of work unit records.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual generation of work unit records from audio and video recordings is used, then users can create records with human judgment and context understanding, but the process is time-consuming and decreases workflow efficiency

Engineering Contradiction:

ImproveaccuracyVSAvoidworkflow efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system enables self-service by automatically generating work unit records from uploaded audio and video files without requiring manual transcription or data entry. The machine learning model processes the media files autonomously to extract and structure relevant information into record fields, eliminating the need for users to manually create records while maintaining accuracy through intelligent content recognition.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical manual process of creating records with an automated machine learning system. Instead of users manually transcribing audio/video content and filling in record fields, the system uses trained ML models to automatically process media files, extract meaningful information, and populate record structures, thereby substituting human labor with automated intelligent processing.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Manufacturing precision

If manual generation of work unit records is used, then users can provide precise definitions of information, but the process is prone to user error

Engineering Contradiction:

Improveprecision of record informationVSAvoidaccuracy

Core Design Contradiction:

Manufacturing precisionVSReliability

Solution Approach 1:

The system incorporates feedback mechanisms where the machine learning model processes uploaded media files and generates structured record information, which can then be reviewed and refined by users. The model learns from corrections and adjustments made to generated records, improving its accuracy over time while maintaining precise information extraction through iterative refinement and validation processes.

Inventive Principle:

Principle #23Feedback

3Productivity

If automated record generation from audio and video recordings is implemented, then workflow efficiency is improved, but the system complexity increases

Engineering Contradiction:

Improveworkflow efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system segments the complex automated record generation process into distinct functional modules: audio/video file upload handling, machine learning model processing, information extraction, record structure generation, and user interface presentation. This segmentation allows each component to be developed, maintained, and optimized independently while working together to achieve efficient automated record creation from media files.

Inventive Principle:

Principle #1Segmentation

4Reliability

If automated record generation is implemented, then human intervention and errors are reduced, but the initial setup and training requirements increase

Engineering Contradiction:

Improveerror reductionVSAvoidtraining and setup time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-training machine learning models on relevant data before deployment. The models are pre-configured with domain-specific knowledge and patterns related to work unit records, allowing them to accurately process uploaded audio and video files immediately upon use. This preliminary preparation minimizes the need for extensive setup and training time when the system is actually deployed for automated record generation.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12619954B2Systems and methods to generate records within a collaboration environment based on a machine learning model trained from a text corpus

Publication Date: 2026.05.05 ASANA INC

AI summary

Systems and methods to generate records within a collaboration environment are described herein. Exemplary implementations may perform one or more of: manage environment state information maintaining a collaboration environment; effectuate presentation of a user interface through which users upload digital assets representing recorded audio and/or video content; obtain input information defining the digital assets input via the user interface; generate transcription information characterizing the recorded audio and/or video content of the digital assets; provide the transcription information as input into a trained machine-learning model; obtain the output from the trained machine-learning model, the output defining one or more new records based on the transcripts; and/or other operations.