Voice Recognition State Management for Continuous Query Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices require a repetitive wake-up operation for each voice command, inconveniently necessitating a wake-up voice before each query, which disrupts continuous user interactions and slows down the processing of successive queries.
Innovation Solution
An electronic apparatus with a microphone, speaker, and processors that identify a wake-up voice to transition from a standby to a wake-up state, allowing continuous voice recognition and query processing by maintaining the wake-up state if subsequent queries exhibit high semantic similarity with previously identified queries, based on user context information and query prediction models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a wake-up operation is performed before each voice command, then the device can accurately interpret user voice without misinterpreting other utterances, but the user interaction becomes repetitive and inefficient
Solution Approach 1:
The system performs a wake-up operation once at the beginning to enter a wake-up state, preparing the voice interpretation system in advance. This preliminary action eliminates the need for repeated wake-up operations before each query, thereby improving processing speed while maintaining interpretation accuracy through the initially activated voice recognition system.
Solution Approach 2:
The voice interpretation system maintains a continuous wake-up state after the initial wake-up operation, allowing uninterrupted processing of multiple queries. This continuous operation eliminates repetitive wake-up cycles, improving productivity while ensuring reliable voice interpretation throughout the sustained active period.
2Productivity
If the device maintains wake-up state continuously, then continuous voice queries can be processed without repetition, but energy consumption increases
Solution Approach 1:
The system dynamically adjusts its operational state based on query characteristics. It maintains a wake-up state for processing continuous queries with sufficient semantic similarity, but can transition to a lower-power state when queries fall outside the expected semantic range, thus balancing continuous processing capability with energy conservation.
Solution Approach 2:
The system changes operational parameters by adjusting the wake-up state duration and activation thresholds based on query patterns. When semantic similarity thresholds are met, the system maintains the wake-up state; when thresholds are not met, it transitions states, thereby optimizing energy consumption while preserving continuous processing capability for relevant queries.
3Reliability
If semantic similarity threshold is set high, then irrelevant queries are filtered out, but legitimate queries may be rejected
Solution Approach 1:
The system performs preliminary voice processing and semantic analysis to evaluate query relevance before applying the similarity threshold. This preliminary action provides context information that helps adjust threshold application, ensuring that legitimate queries are not incorrectly rejected while maintaining reliable filtering of irrelevant queries.
Solution Approach 2:
The system uses feedback from query analysis results to dynamically adjust threshold application. When semantic similarity is evaluated, the system considers contextual feedback to determine whether to accept or reject queries, thereby maintaining high relevance filtering while preserving adaptability to legitimate variations in user queries.
Data Source
AI summary
An electronic apparatus includes a microphone; a speaker; a memory storing at least one instruction; and one or more processors operatively coupled to the memory and the speaker, wherein the one or more processors are configured to execute the at least one instruction to: based on identifying that a first voice sensed through the microphone corresponds to a wake-up voice, convert a state of the electronic apparatus from a standby state to a wake-up state, based on a second voice being sensed while the state of the electronic apparatus is the wake-up state, identify a first user query included in first voice data obtained based on the second voice, perform a first operation corresponding to the first user query, identify a predetermined query corresponding to the first user query, obtain a query list including at least one query based on the predetermined query and user context information, identify a second user query included in second voice data obtained based on a third voice sensed through the microphone, and based on a first semantic similarity between the second user query and the at least one query of the query list being greater than or equal to a predetermined value, perform a second operation corresponding to the second user query and maintain the state of the electronic apparatus as the wake-up state.