Speech Correction System with Local Vocabulary Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional voice systems require frequent user intervention to modify and improve language analysis models, causing inconvenience by necessitating repeated confirmation and retraining processes.
Innovation Solution
A speech correction system comprising a storage device, audio receiver, and processing device with a speech recognition engine and determination module that identifies key speech patterns, generates candidate vocabulary lists with scores, and stores vocabularies or transcodes based on score thresholds, allowing for local storage and quick recognition without cloud server comparison.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional cloud-based voice systems are used to perform speech recognition and language analysis, then speech recognition capability is provided, but user convenience deteriorates due to frequent confirmation requests and retraining requirements
Solution Approach 1:
The patent segments the speech recognition system into multiple components: a speech recognition engine that processes audio signals, a determination module that evaluates vocabulary scores, and a database that stores recognized speech patterns. This segmentation allows the system to autonomously handle recognition tasks without requiring centralized cloud processing for every interaction, thereby improving user convenience while maintaining recognition reliability
Solution Approach 2:
The system implements self-service by automatically determining whether to store candidate vocabularies or transcodes based on vocabulary scores, without requiring user confirmation. The determination module autonomously evaluates recognition results and makes storage decisions, eliminating the need for users to repeatedly confirm or retrain the system, thus resolving the contradiction between maintaining recognition capability and improving ease of operation
2Measurement precision
If cloud server-based language analysis is used to determine vocabulary accuracy, then speech analysis accuracy is improved, but system complexity and response time worsen due to network dependency
Solution Approach 1:
The system performs preliminary speech pattern recognition and vocabulary scoring locally using the speech recognition engine before any potential cloud-based verification. By pre-processing audio signals and generating candidate vocabularies with confidence scores locally, the system reduces the need for complex cloud-based analysis, thereby improving response time while maintaining necessary accuracy through the determination module's score-based decision-making
Solution Approach 2:
The patent extracts the essential speech recognition and evaluation functions from the cloud-based system and implements them locally through the speech recognition engine and determination module. This extraction allows the system to operate independently of cloud servers for routine operations, reducing system complexity and network dependency while maintaining vocabulary analysis accuracy through local score evaluation
3Measurement precision
If user confirmation is required for every speech recognition result, then recognition accuracy is verified, but processing efficiency deteriorates due to repeated interactions
Solution Approach 1:
The system changes the parameter of user interaction from mandatory confirmation to conditional storage based on vocabulary scores. The determination module evaluates the vocabulary score parameter and automatically decides whether to store the candidate vocabulary or transcode in the database. This parameter-based decision-making maintains recognition accuracy verification through score evaluation while dramatically improving processing efficiency by eliminating repeated user confirmations
Data Source
AI summary
The speech correction system includes a storage device, an audio receiver and a processing device. The processing device includes a speech recognition engine and a determination module. The storage device is configured to store a database. The audio receiver is configured to receive an audio signal. The speech recognition engine is configured to identify a key speech pattern in the audio signal and generate a candidate vocabulary list and a transcode corresponding to the key speech pattern; wherein the candidate vocabulary list includes a candidate vocabulary corresponding to the key speech pattern and a vocabulary score corresponding to the candidate vocabulary. The determination module is configured to determine whether the vocabulary score is greater than a score threshold. If the vocabulary score is greater than the score threshold, the determination module stores the candidate vocabulary corresponding to the vocabulary score in the database.