Dynamic Voice Grammar for Search Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Voice recognition systems face reliability issues due to the inverse proportionality between grammar size and recognition accuracy, making it challenging to provide reliable searches for large databases, as large grammars are impractical and omitting terms results in unrecognized valid queries.
Innovation Solution
A dynamic grammar is generated based on user input, narrowing the search scope by prompting users to enter a set of characters, which reduces the grammar size and improves recognition accuracy, and allows for refinement of searches by generating grammars from initial search results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If all or significant portion of terms in the searchable domain are incorporated into the grammar, then the system can recognize more valid queries, but the grammar becomes too large to provide reliable voice recognition
Solution Approach 1:
The patent segments the grammar into multiple versions, each corresponding to a different subset of database items. Instead of using a single comprehensive grammar containing all terms, the system divides the vocabulary into manageable segments based on search context and dynamically selects or generates the appropriate segment for each query, thereby maintaining both coverage and recognition reliability.
Solution Approach 2:
The patent employs dynamic grammar generation where the grammar is not fixed but adapts based on the search query context. The system dynamically determines which subset of terms to include in the grammar based on the current search state, user input, and database structure, allowing the grammar to evolve and adjust its size and content to optimize recognition accuracy for each specific search scenario.
2Reliability
If many terms are omitted from the grammar to maintain small size, then recognition reliability improves, but the system becomes incapable of recognizing many valid queries
Solution Approach 1:
The patent creates a universal search system that can handle multiple search scenarios through a single integrated framework. The grammar generation mechanism is designed to be universally applicable across different search contexts, product categories, and query types, while dynamically adapting the grammar content to ensure both comprehensive coverage and high recognition accuracy for any given search scenario.
3Adaptability or versatility
If a large grammar is used to cover all database items, then more terms can be recognized, but misinterpretations by the voice recognition system increase
Solution Approach 1:
The patent applies local quality by tailoring the grammar to specific local contexts rather than using a generic comprehensive grammar. For each search query, the system identifies the relevant subset of terms and generates a localized grammar that includes only those terms, thereby reducing confusion and misinterpretations while maintaining accurate recognition for the specific search context at hand.
Data Source
AI summary
A system and associated methods are disclosed for improving voice recognition accuracy when a user conducts a search by voice. One method involves prompting the user to enter a set of characters of the query (e.g., the first N letters of a query term), and then using these letters to execute a preliminary search. The results of the preliminary search are then used to generate a dynamic grammar for interpreting the full voice query. The grammar may alternatively be retrieved from a cache or other memory that stores the grammars for various combinations of letters. In one embodiment, the user enters the characters by selecting the corresponding keys on a standard telephone keypad (one depression per letter) and then saying the letters, and the keypad entries are used to reduce the number of possible interpretations of each character utterance. Another method, which is useful for search refinement, involves generating a dynamic grammar from a set of search results (e.g., when the number of hits is large), and then using this grammar to interpret utterances of additional query terms to be added to the query.