Speech Recognition System Self-Service Model Update
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated speech recognition systems require frequent updates and human expertise to improve performance, which is time-consuming and labor-intensive, and face challenges in collecting sufficient data for accurate model building.
Innovation Solution
A method where a computer-based speech recognition system processes speech utterances using internal representations, compares its tasks with human-performed tasks, and updates these representations to enhance recognition performance, using modules like the comparator and IR update determination modules to adjust parameters automatically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If human experts manually review and modify internal representations to improve speech recognition performance, then accuracy is improved, but time consumption and labor intensity increase
Solution Approach 1:
The system performs self-updating of internal representations by automatically comparing its task outputs with human task outputs and modifying its own acoustic model parameters without requiring manual expert intervention, thereby reducing time loss while maintaining accuracy improvement
Solution Approach 2:
The system implements a feedback mechanism where human task outputs serve as reference data to compare against computer-performed tasks, and this comparison feedback drives automatic modifications of internal representations, enabling continuous accuracy improvement without proportional increase in human time investment
2Measurement precision
If human experts manually modify internal representations to improve speech recognition performance, then accuracy is improved, but labor intensity increases
Solution Approach 1:
The system automatically performs the complex operation of modifying internal representations by comparing its outputs with human outputs and self-correcting its acoustic model parameters, eliminating the need for human experts to perform labor-intensive manual adjustments while maintaining accuracy improvements
Solution Approach 2:
The patent replaces the mechanical process of manual expert analysis and modification with an automated computational process that uses algorithmic comparison and automatic parameter adjustment, thereby simplifying operation while achieving the same accuracy improvement goals
3Measurement precision
If sufficient data is collected to build accurate speech recognition models, then model accuracy is improved, but data collection time increases
Solution Approach 1:
The system performs preliminary automatic model updates using available data before extensive data collection is complete, comparing its outputs with human outputs and adjusting its acoustic model parameters in advance, thereby achieving improved accuracy without waiting for complete data collection
Solution Approach 2:
The system enables continuous model improvement through ongoing automatic comparisons with human outputs and continuous parameter adjustments, eliminating the interruption caused by data collection pauses and maintaining steady accuracy improvement over time
Data Source
AI summary
Systems and methods for modifying a computer-based speech recognition system. A speech utterance is processed with the computer-based speech recognition system using a set of internal representations, which may comprise parameters for recognizing speech in a speech utterance, such as parameters of an acoustic model and/or a language model. The computer-based speech recognition system may perform a first task in response to the processed speech utterance. The utterance may also be provided to a human who performs a second task based on the utterance. Data indicative of the first task, performed by the computer system, is compared to data indicative of a second task, performed by the human in response to the speech utterance. Based on the comparison, the set of internal representations may be updated or modified to improve the speech recognition performance and capabilities of the speech recognition system.