Speech Recognition Phrase Reutilization for Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current voice user interface (VUI) testing techniques are inefficient and costly due to the need for version-specific recordings and manual selection of phrases for speech recognition grammar testing, leading to significant storage and recording costs, as well as suboptimal phrase coverage.
Innovation Solution
A system and method that maintains a set of audio phrases with associated characteristics in a database, using a coverage tool to compare textual phrases against translations to determine phrase coverage for finite state grammars, allowing for the identification of valid test phrases and automatic notification of coverage gaps, thereby optimizing phrase usage and reducing recording costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If version-specific test sets are created for each VUI version, then speech recognition grammar testing accuracy is improved, but storage costs and recording costs increase significantly
Solution Approach 1:
The patent creates a universal phrase repository that serves multiple VUI versions simultaneously. Instead of creating separate test sets for each version, a single repository of audio phrases with associated metadata (transcriptions, grammars, VUI versions) is maintained. This repository can be queried to generate appropriate test sets for any specific VUI version, eliminating the need for duplicate storage and reducing recording costs while maintaining testing accuracy.
Solution Approach 2:
The patent merges multiple version-specific test sets into a single integrated phrase repository. By combining audio phrases from different versions and associating them with metadata indicating which versions they apply to, the system consolidates storage requirements. The repository structure allows efficient retrieval of relevant phrases for any specific version without requiring separate physical copies.
2Quantity of substance
If manual selection of phrases from external sources is performed, then recording costs are reduced, but time consumption and complexity increase
Solution Approach 1:
The patent implements self-service through automated query processing. The system automatically queries the phrase repository based on specified criteria (VUI version, grammar, phrase coverage requirements) and generates complete test sets without manual intervention. The automated processing includes filtering, selecting, and organizing phrases according to the query parameters, eliminating the need for manual phrase selection while maintaining cost efficiency.
Solution Approach 2:
The patent incorporates feedback mechanisms where the system evaluates the results of phrase selection and automatically adjusts subsequent queries. The coverage analysis provides feedback on which phrases have been selected and which are missing, allowing the system to iteratively improve test set completeness. This automated feedback loop replaces manual review and selection processes, reducing both time consumption and complexity.
3Reliability
If comprehensive phrase coverage is achieved for all VUI versions, then testing completeness is improved, but system complexity increases
Solution Approach 1:
The patent segments the testing system into distinct modular components: a phrase repository storing audio phrases with metadata, a query processing module that accepts testing requirements, a phrase selection module that retrieves appropriate phrases, and a coverage analysis module that evaluates completeness. Each component has a specific function and can be independently maintained or modified. This segmentation manages system complexity by creating clear boundaries and responsibilities for each module while achieving comprehensive coverage through their coordinated operation.
Data Source
AI summary
A set of audio phrases and corresponding phrase characteristics can be maintained, such as in a database. The phrase characteristics can include a translation of speech in the associated audio phrase. A finite state grammar that includes a set of textual phrases can be received. A software algorithm can execute to compare the set of textual phrases against the translations associated with the maintained audio phrases. A result of the software algorithm execution can be produced, where the result indicates phrase coverage for the finite state grammar based upon the audio phrases.