Combined Speech Profile for On-Device Multi-User Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional speech recognition systems for multiple users are not optimized for on-device processing and require frequent interactions with servers to handle the computing resources needed for speech recognition across multiple speech profiles, limiting their efficiency and effectiveness.
Innovation Solution
A system that combines multiple speech profiles to generate a single speech recognition result, allowing for on-device processing of speech input from multiple users by selecting a relevant recognition result based on an identified voice profile, thereby reducing the need for server interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple speech profiles are used for speech recognition, then speech recognition accuracy for multiple users is improved, but device complexity and server interaction frequency increase
Solution Approach 1:
The patent combines multiple speech profiles into a single combined speech profile that can be used for speech recognition across multiple users. This merging approach maintains the ability to recognize speech from different users while reducing system complexity by eliminating the need to manage and switch between multiple separate speech profiles.
Solution Approach 2:
The combined speech profile serves multiple functions by accommodating speech recognition for multiple different users simultaneously. This universal profile acts as a multi-functional solution that replaces the need for user-specific profiles, thereby simplifying the system while maintaining recognition accuracy.
2Measurement precision
If multiple speech profiles are processed, then speech recognition results for different users are improved, but processing time and server interactions increase
Solution Approach 1:
By merging multiple speech profiles into one combined profile, the system processes speech input in a single operation rather than requiring separate processing for each user profile. This eliminates the time loss associated with switching between profiles and reduces the need for multiple server interactions.
Solution Approach 2:
The combined speech profile is prepared in advance to encompass multiple users' speech characteristics, so when speech input is received, the system can immediately process it without needing to retrieve or switch between individual profiles, thereby reducing processing time.
3Adaptability or versatility
If traditional speech recognition systems are used, then speech input from multiple users can be handled, but on-device processing capability is reduced and server reliance increases
Solution Approach 1:
The patent merges multiple speech profiles into a single combined profile that can be stored and processed locally on the device. This enables the device to autonomously handle speech recognition for multiple users without requiring constant server interactions, thereby improving on-device processing capability while maintaining multi-user versatility.
Data Source
AI summary
Systems and processes for speech recognition for multiple users are provided. For example, in response to receiving speech input from a user, a combined speech profile is obtained from a plurality of speech profiles. The speech input is interpreted based on the combined speech profile to obtain a plurality of speech recognition results. The plurality of speech recognition results includes a first speech recognition result corresponding to a first speech profile of the plurality of speech profiles, wherein the first speech profile corresponds to a first user, and a second speech recognition result corresponding to a second speech profile of the plurality of speech profiles, wherein the second speech profile corresponds to a second user different from the first user. A respective speech recognition result based on an identified voice profile is then selected from the plurality of speech recognition results.