Speech Prompt Query for Smart Device Interaction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Smart devices require high user dependency for speech interaction, leading to low probability of correct speech instruction input and uncontrolled situations due to the lack of real-time speech prompts during the interaction process.

Innovation Solution

A speech interactive method and device that detect user operations during application interface display, query and set applicable speech instructions, and provide prompts based on these instructions to facilitate real-time speech interaction, reducing user dependency and guiding users through available functions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If the smart device provides speech prompts only after user input, then the device complexity is reduced, but the user dependency increases and speech recognition failure probability increases

Engineering Contradiction:
Improvedevice complexityVSAvoidspeech recognition failure probability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The system queries and displays applicable speech instructions before the user inputs speech, during the interface display process. This preliminary provision of speech prompts enables users to know what instructions are available before attempting speech input, reducing recognition failure probability without significantly increasing device complexity

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system provides real-time feedback by displaying speech prompts that are relevant to the current interface state. This feedback loop allows users to adjust their speech input based on the displayed prompts, improving speech recognition success rate while maintaining manageable device complexity

Inventive Principle:
Principle #23Feedback

2Reliability

If the smart device provides comprehensive speech prompts in advance, then the speech recognition success rate improves, but the device complexity increases

Engineering Contradiction:
Improvespeech recognition success rateVSAvoiddevice complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

Instead of providing all possible speech prompts uniformly, the system queries and displays only the speech instructions that are locally relevant to the current interface and application state. This selective provision of prompts improves recognition success rate for context-appropriate commands while avoiding the complexity of managing comprehensive prompt databases

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The speech prompt system is segmented into application-specific instructions and system general instructions. The query process separately retrieves relevant prompts from different sources based on the current context, reducing the complexity of managing a single comprehensive prompt system while maintaining high recognition success rates

Inventive Principle:
Principle #1Segmentation

3Ease of operation

If the user must remember speech prompts in advance, then the ease of operation decreases, but the information processing load on the device decreases

Engineering Contradiction:
Improveease of operationVSAvoidinformation processing load
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The system automatically queries and displays the relevant speech instructions based on the current interface state, without requiring users to remember prompts in advance. This self-service approach improves ease of operation by providing context-relevant prompts automatically, while the information processing load is managed through selective querying based on interface context

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS11238858B2Speech interactive method and device
Publication Date: 2022.02.01 BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
  • US11238858B2 patent drawing
  • US11238858B2 patent drawing
  • US11238858B2 patent drawing

AI summary

The present disclosure provides a speech interactive method and device. The method includes: during a process of displaying an interface of an application, detecting a first user operation; in response to the first user operation, querying a speech instruction associated with the application based on the application containing the interface and querying a system general speech instruction; setting the speech instruction and the system general speech instruction as applicable speech instructions for the application; and providing a prompt based on the applicable speech instructions to perform a speech interaction.