AI Speech Recognition with Alternative Result Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional speech recognition systems fail to recognize misrecognition errors, leading to user dissatisfaction as they do not generate alternative recognition results without requiring the user to re-utter their speech, resulting in inefficient interactions.
Innovation Solution
An artificial intelligence apparatus and method that acquires speech data, generates an initial recognition result, determines incorrect recognition based on user feedback, corrects word probabilities, and generates an alternative recognition result without requiring re-utterance, allowing for improved accuracy and user satisfaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional speech recognition systems perform control based on initial recognition results, then the system operates efficiently with simple processing, but the system fails to recognize misrecognition errors and requires user re-utterance
Solution Approach 1:
The system performs preliminary analysis of multiple alternative recognition results before final execution. When negative feedback is detected, the system has already prepared alternative interpretations of the speech data, allowing it to quickly switch to a correct interpretation without requiring user re-utterance, thus improving reliability while maintaining reasonable complexity
Solution Approach 2:
The system implements a feedback mechanism where negative feedback from the user triggers re-processing of the speech data. The system generates alternative recognition results based on the same speech data and user feedback, allowing it to correct misrecognition errors and improve reliability through iterative refinement
2Loss of time
If the system generates alternative recognition results upon negative feedback, then user satisfaction improves and re-utterance is reduced, but processing time and computational resources increase
Solution Approach 1:
The system generates multiple alternative recognition results in advance, before user feedback is received. When negative feedback occurs, the system can immediately present alternative interpretations without requiring additional processing time or user re-utterance, thus reducing time loss while managing computational resources efficiently
Solution Approach 2:
The system performs partial re-processing only when negative feedback is detected, rather than continuously processing all speech data. It generates alternative results selectively based on the need, balancing the reduction of user re-utterance time with maintained processing efficiency
Data Source
AI summary
An artificial intelligence apparatus for recognizing speech of a user includes a microphone and a processor configured to acquire, via the microphone, first speech data including speech of a user, generate a first speech recognition result corresponding to the first speech data, perform control corresponding to the generated first speech recognition result, generate an alternative speech recognition result corresponding to the first speech data if negative feedback is acquired from the user, and perform control corresponding to the generated alternative speech recognition result.