Speech Interface for Search Engine Query Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines do not offer a functional solution for users to make queries by speaking conversational words into a voice or speech recognition system, as existing speech recognition technologies are limited in understanding human speech complexities and are not compatible with various speech recognition methods, restricting their use in search engine queries.
Innovation Solution
A system that receives voiced utterances, converts them into data strings, identifies a user-chosen search engine, modifies query terms by replacing spaces with compatible separators, constructs a uniform resource locator, and submits the query to the chosen search engine, enabling speech-initiated searches across different search engines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If speech recognition technology is used for search engine queries, then user interaction convenience is improved, but compatibility with various speech recognition methods deteriorates
Solution Approach 1:
The patent introduces a speech recognizer as an intermediary component that converts spoken words into text queries. This mediator handles the complexity of different speech recognition methods internally, allowing the search engine to work with various speech recognition systems without requiring direct integration with each specific method. The intermediary abstracts away the differences between speech recognition approaches while maintaining user interaction convenience.
2Productivity
If speech recognition systems are used to convert spoken words to data, then query input efficiency is improved, but accuracy in understanding human speech complexities deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the system provides feedback to the user about the accuracy of speech recognition. When recognition errors occur, the system can ask for clarification or allow the user to correct the query. This feedback loop helps maintain high accuracy in understanding human speech complexities while preserving the efficiency benefits of speech-based query input.
3Device complexity
If limited keyword sets are programmed into speech recognition systems, then system simplicity is improved, but search query capability deteriorates
Solution Approach 1:
The patent makes the speech recognition system universal by enabling it to handle not only pre-programmed keywords but also conversational words and phrases. The system can recognize a wide variety of speech patterns and convert them into appropriate search queries, making it multi-functional for different types of search needs while maintaining relative simplicity through a unified processing approach.
Data Source
AI summary
An embodiment provides search results from a speech initiated search query. The system receives voiced utterances from a user, converts the voiced utterances through use of a speech recognition application, system or method into data strings, identifies from the data strings a search engine identifier representing a search engine chosen by the user to perform a search, identifies from the data strings a query term to be searched for, modifies the query term to be searched for by replacing any spaces in the query term to be searched for with a query term separator compatible with the search engine represented by the search engine identifier thereby creating a modified query term to be searched for, constructs a uniform resource locator that includes the modified query term to be searched for such that the constructed uniform resource locator represents a valid request to the search engine represented by the search engine identifier, opens the constructed uniform resource locator which causes the chosen search engine to make a search for the modified query term, and provides the results of the search system to the user.