Smart Transcription Proxy Deduplication for Audio Streams

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In public safety communication systems, duplicate transcription requests for the same audio stream are sent to transcription services, leading to unnecessary fees and wasteful use of processing resources due to redundant requests from fully redundant site servers that operate independently.

Innovation Solution

A smart transcription proxy service that receives requests from site servers, determines duplicate requests, and selectively sends the audio stream to the transcription service based on random, round-robin, or network parameter-based selection to ensure single transcription of each call, thereby reducing redundant processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If multiple site servers independently forward audio streams to transcription services, then system redundancy and fault tolerance are maintained, but duplicate transcription requests are generated causing unnecessary fees and resource wastage

Engineering Contradiction:

Improvesystem redundancy and fault toleranceVSAvoidprocessing resources and transcription fees

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The patent introduces a proxy server as an intermediary between site servers and transcription services. The proxy server receives transcription requests from multiple site servers, identifies duplicate audio streams using unique identifiers, and forwards only unique streams to the transcription service. This mediator approach maintains system redundancy while eliminating duplicate transcriptions, thus preserving reliability without wasting processing resources.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent combines multiple transcription requests into a single request when they refer to the same audio stream. By merging duplicate requests at the proxy server level and using a unified transcription identifier, the system maintains the appearance of multiple independent sites while actually performing transcription only once, thereby reducing resource consumption while preserving system redundancy.

Inventive Principle:

Principle #5Merging (Combining)

2Productivity

If each site server independently requests transcription, then distributed processing and fault tolerance are achieved, but redundant transcription requests increase processing overhead and costs

Engineering Contradiction:

Improvedistributed processing capabilityVSAvoidtranscription request management

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The proxy server acts as a mediator that simplifies transcription request management. Site servers continue to operate independently with full transcription request capability, maintaining distributed processing architecture. The proxy server handles the complexity of deduplication and coordination, reducing the management burden on individual site servers while preserving their independent operation capability.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent segments the transcription management function into two parts: site servers generate transcription requests independently (maintaining distributed processing), while the proxy server handles deduplication and coordination (reducing complexity). This segmentation allows each component to focus on its specific task, simplifying overall system management while preserving distributed processing capabilities.

Inventive Principle:

Principle #1Segmentation

3Loss of information

If all transcription requests are processed, then complete transcription coverage is ensured, but duplicate transcriptions waste fees and processing resources

Engineering Contradiction:

Improvetranscription coverageVSAvoidtranscription fees and processing resources

Core Design Contradiction:

Loss of informationVSLoss of energy

Solution Approach 1:

The patent implements a feedback mechanism where the proxy server tracks transcribed audio streams using unique identifiers and informs site servers about duplicate requests. This feedback loop ensures that transcription coverage is maintained (all unique streams are transcribed) while eliminating redundant processing (duplicates are identified and skipped), thus preventing information loss without wasting resources.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The proxy server performs preliminary deduplication checking before forwarding transcription requests to the transcription service. By identifying and filtering duplicate audio streams in advance using unique identifiers, the system ensures that only unique streams are transcribed, maintaining complete transcription coverage while avoiding wasteful duplicate processing and associated fees.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20230403315A1System and method for smart transcription proxy service for limiting audio processing

Publication Date: 2023.12.14 MOTOROLA SOLUTIONS INC

AI summary

Techniques for transcription of an audio stream are provided. A request for transcription of a first audio stream from a first device is received. A request for transcription of a second audio stream from a second device is received. It is determined that the first and second audio streams are the same. One of the first and second audio streams are selected for transcription. Transcription of the selected audio stream is requested.