Voice Query Search Prefetching via Intermediate Transcription
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current speech recognition systems face delays in providing search results as they wait for the full transcription of a voice query before generating results, which can prolong the time taken to return relevant information to the user.
Innovation Solution
The system identifies preliminary search results based on an intermediate transcription of the user's utterance, allowing it to provide results before the full transcription is complete by associating the initial portion of the query with past queries and retrieving cached results, thereby reducing the time to return search results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system waits for the full transcription of a voice query before generating search results, then the accuracy of search results is improved, but the time taken to return results to the user increases
Solution Approach 1:
The system performs preliminary search using intermediate transcriptions of the voice query before the full transcription is available. This allows search results to be generated and returned to the user in advance, reducing the overall response time while maintaining accuracy through subsequent refinement when the full transcription becomes available
2Reliability
If the system processes the complete voice query before providing results, then the reliability of result matching is improved, but the responsiveness of the system deteriorates
Solution Approach 1:
The voice query processing is segmented into multiple stages: initial intermediate transcription processing that provides preliminary results quickly, followed by subsequent processing of the full transcription to refine and verify the results. This segmentation allows the system to deliver responsive preliminary results while maintaining reliable final matching
Data Source
Figure 1a
Figure 1b
Figure 2
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for receiving audio data encoding an utterance; obtaining an intermediate transcription of the utterance; before a final transcription of the utterance is obtained: i) determining that the intermediate transcription of the utterance is associated with a previously received search query, ii) obtaining one or more results that are identified as responsive to the previously received search query, and iii) storing one or more of the results; obtaining the final transcription of the utterance; determining that the final transcription of the utterance is also associated with the previously received search query; and in response to determining that the final transcription of the utterance is also associated with the previously received search query, providing the stored one or more results for output.