Speech Recognition Service Early Partial Response Mechanism

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The response time for electronic devices to speech inputs is prolonged, leading to user discomfort and reduced efficiency in speech recognition services due to the need for complete processing before providing a response.

Innovation Solution

Implementing a method where the electronic device provides an early partial response to the user's speech input, allowing the device to output a first response upon completing initial processing, followed by subsequent responses once the intent is fully derived and additional information is received from servers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the electronic device waits for complete processing of speech input before providing response, then the accuracy and completeness of the response is improved, but the response time increases causing user discomfort

Engineering Contradiction:
Improveresponse accuracyVSAvoidresponse wait time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The electronic device performs preliminary processing actions on the speech input as soon as they are completed, generating and outputting a first response based on initial processing results (such as speech-to-text conversion) before the complete intent derivation is finished. This preliminary action reduces the user's perceived wait time while the background processing continues to generate more accurate subsequent responses.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The response generation process is segmented into multiple stages: initial processing (speech-to-text), intent derivation, and information retrieval. Each stage can independently generate a response portion, allowing the system to provide incremental responses rather than waiting for complete processing. This segmentation enables the first response to be provided early while maintaining options for subsequent refined responses.

Inventive Principle:
Principle #1Segmentation

2Loss of time

If the electronic device provides early partial response before complete processing, then the user wait time is reduced, but the completeness and accuracy of the response may be compromised

Engineering Contradiction:
Improveresponse wait timeVSAvoidresponse accuracy
Core Design Contradiction:
Loss of timeVSReliability

Solution Approach 1:

The system performs preliminary processing actions (speech-to-text conversion, basic intent identification) as soon as they are completed, generating a first response that provides immediate feedback to the user. This preliminary action is designed to be less accurate than complete processing but sufficient to reduce user wait time and provide initial information.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The response is segmented into multiple levels of completeness: a first response based on initial processing that provides partial information quickly, and subsequent responses that build upon or correct the initial response as more processing is completed. This segmentation allows the system to trade off between response speed and accuracy at different stages.

Inventive Principle:
Principle #1Segmentation

3Reliability

If the electronic device processes speech input completely before responding, then the service reliability is maintained, but the operating efficiency decreases

Engineering Contradiction:
Improveservice reliabilityVSAvoidoperating efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The electronic device performs preliminary processing and generates a first response as soon as initial processing is complete, rather than waiting for full processing. This preliminary action increases operating efficiency by providing responses faster, while the system maintains service reliability by continuing background processing to generate subsequent, more accurate responses if needed.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts its response strategy based on processing progress and user needs. It can provide incremental responses at different stages of processing, allowing the operating efficiency to improve through faster initial responses while maintaining reliability through optional subsequent refinements. The level of processing completeness becomes a dynamic parameter rather than a fixed requirement.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3642838B1Method for operating speech recognition service and electronic device and server for supporting the same
Publication Date: 2023.11.15 SAMSUNG ELECTRONICS CO LTD
  • EP3642838B1 patent drawingFigure 1
  • EP3642838B1 patent drawingFigure 2
  • EP3642838B1 patent drawingFigure 3a

AI summary

Disclosed is an electronic device including a communication circuit, a microphone, a memory, speaker, and a processor, in which the processor receives the speech input and transmits first data associated with the speech input to a first server for supporting the speech recognition service, receives second data corresponding to processing of a part of the first data from the first server and outputs the second data at a first time that a first period of time has elapsed after the transmission of the first data, and outputs third data corresponding to processing of the rest of the first data at a second time that a second period of time has elapsed from the first time, while receiving the third data from the first server or a second server for supporting the speech recognition service before the second time.