Audio Cache Matching for Accurate Voice Query Execution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Speech recognition engines in television control devices suffer from accuracy issues and incorrect operations, particularly with common voice queries, leading to frustrating user experiences and increased network traffic due to unnecessary speech recognition processing.
Innovation Solution
An audio cache system that processes common voice queries without speech recognition and selectively disables incorrect cache entries based on user feedback, sending ambiguous queries to speech recognition processors to improve accuracy and reduce network traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If speech recognition processing is used for all voice queries, then accuracy is improved, but network traffic increases and processing time increases
Solution Approach 1:
The patent segments voice queries into two categories: common queries handled by audio cache and ambiguous queries handled by speech recognition processing. This segmentation allows the system to avoid unnecessary speech recognition for common queries, reducing network traffic while maintaining accuracy for ambiguous cases.
Solution Approach 2:
The patent implements preliminary action by pre-storing audio samples of common voice queries in an audio cache before they are received. This allows the system to quickly match incoming queries against the cache without immediately invoking speech recognition processing, reducing both network traffic and processing time for common queries.
2Productivity
If audio cache is used for common voice queries, then network traffic is reduced and processing speed is improved, but accuracy decreases due to incorrect matching
Solution Approach 1:
The patent implements feedback by monitoring user responses to cache matches. When a user corrects a misidentified query or indicates the operation was incorrect, the system learns from this feedback and adjusts future matching behavior, improving accuracy while maintaining the speed benefits of caching.
Solution Approach 2:
The patent introduces an intermediary verification step between audio cache matching and execution. Ambiguous matches are routed to speech recognition processing for verification before execution, ensuring accuracy while allowing direct execution for clear matches to maintain speed.
3Ease of operation
If cache entries are disabled for individual users to prevent errors, then user experience is improved, but network traffic increases for that user
Solution Approach 1:
The patent applies local quality by customizing the audio cache behavior for each user based on their individual error patterns. Users who frequently experience incorrect matches have their problematic cache entries disabled, while other users continue to benefit from the cache. This localized approach improves individual user experience without globally disabling the cache.
Solution Approach 2:
The patent changes the operational parameters of cache entries dynamically based on user feedback and error rates. Cache entries can be enabled, disabled, or adjusted in confidence thresholds based on accumulated data, allowing the system to optimize both user experience and network traffic for each individual user.
Data Source
AI summary
An audio file associated with a user voice query may be received at a user device. The audio file may be compared to a plurality of references, such as cache entries, corresponding to a plurality of other voice queries. Based on a determination that the voice query corresponds to one of the references, an operation associated with the voice query may be executed. An indication may be received that the operation was not an intended operation associated with the voice query. Based on receiving this indication, the incorrectly identified operation, associated reference, e.g., voice query, may be disabled for the user or the device. However, the cache entry may remain enabled for one or more of a plurality of other devices.