Personalized Offline Voice Commands Through Online Grammar Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital assistants are limited in offline functionality, failing to recognize personalized user commands without network connectivity, leading to suboptimal user experience and increased network reliance.
Innovation Solution
A system and method for updating the grammar of a speech recognizer on a user device by connecting to a server, learning personalized commands, and integrating them into the local grammar for offline execution, reducing network dependence and improving command recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the speech recognizer uses a limited offline grammar subset, then offline command recognition is enabled, but personalized user commands cannot be recognized without network connectivity
Solution Approach 1:
The system performs preliminary actions by detecting user utterances during online mode, analyzing them to identify personalized commands, and updating the offline grammar in advance. This preparation phase occurs when network connectivity is available, so that when offline mode is activated, the device already has an expanded grammar capable of recognizing personalized commands without requiring real-time network access.
2Measurement precision
If the system continuously consults the server for command recognition, then accurate command interpretation is achieved, but network bandwidth and battery are consumed
Solution Approach 1:
The system segments command recognition into two distinct modes: online mode for learning and analyzing personalized commands with server assistance, and offline mode for executing recognized commands using the locally stored updated grammar. This segmentation allows the device to achieve accurate command interpretation when needed while minimizing network consultations to only those moments when personalized commands are being learned, thereby reducing bandwidth and battery consumption during normal operation.
3Adaptability or versatility
If the grammar is updated with personalized terms from online usage, then offline command recognition improves, but device storage requirements increase
Solution Approach 1:
The system applies local quality by selectively updating the offline grammar only with specific personalized terms and phrases that the particular user actually uses, rather than storing all possible commands. The grammar expansion is tailored to each user's specific speech patterns and preferred phrasing, making the storage investment efficient by only capturing locally relevant linguistic variations that improve that specific user's offline experience.
Data Source
AI summary
Methods, systems, apparatus, including computer programs encoded on a computer storage medium, for a user device to learn offline voice actions. In one aspect, the method includes actions of detecting, by the user device, an utterance at a first time when the user device is connected to a server by a network, providing, by the user device, the utterance to the server using the network, receiving, by the user device and from the server, an update to the grammar of the user device, detecting, by the user device, a subsequent utterance of the utterance at a second time when the user device is not connected to the server by a network, and in response to detecting, by the user device, the subsequent utterance of the utterance at the second time, identifying, by the user device, an operation to perform based on (i) the subsequent utterance, and (ii) the updated grammar.