Language-Agnostic Command Mapping via Action Datasets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional digital assistants are limited by the need for users to structure their spoken commands in a specific language and dialect, often leading to misinterpretations and inability to execute actions within existing applications on mobile devices, particularly when users are on the move.
Innovation Solution
A digital assistant system that generates and distributes crowd-sourced action datasets, allowing users to invoke actions in their natural language, with a central server translating and mapping commands across various languages and dialects, enabling seamless execution of actions on computing devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If digital assistants use speech recognition and natural language processing to interpret commands, then command understanding capability is improved, but misinterpretation rate increases due to privacy concerns and signal strength issues
Solution Approach 1:
The patent introduces a crowdsourced action dataset as an intermediary between the user's spoken command and the digital assistant's interpretation. Instead of relying solely on the assistant's internal NLP algorithms, the system queries a centralized database populated by community-contributed action datasets, which serve as a mediator to disambiguate commands and improve interpretation accuracy.
Solution Approach 2:
The system implements feedback mechanisms where users can provide feedback on command interpretations and actions. This feedback is used to refine and update the crowdsourced action datasets, creating a continuous improvement loop that enhances command understanding accuracy over time while addressing misinterpretation issues.
2Reliability
If digital assistants require users to structure commands in specific languages and dialects, then command execution reliability is improved, but user convenience deteriorates due to unnatural command structure requirements
Solution Approach 1:
The patent changes the parameter of command language from fixed, assistant-defined languages to flexible, user-preferred languages. By maintaining a core language structure while allowing translation into various languages and dialects, the system preserves execution reliability while improving ease of operation for users speaking different languages.
Solution Approach 2:
The system creates a universal command understanding capability that works across multiple languages and dialects through the crowdsourced action dataset framework. Users can issue commands in their preferred language, and the system translates and executes them universally, eliminating the need for users to learn specific command structures.
3Adaptability or versatility
If digital assistants execute actions within existing applications on mobile devices, then functionality versatility is improved, but system complexity increases due to integration requirements
Solution Approach 1:
The patent segments the action execution functionality into separate action datasets that can be independently created, stored, and executed. Each action is encapsulated as a discrete unit within the crowdsourced database, allowing the system to execute actions from existing applications without requiring complex integration logic, thus reducing system complexity while maintaining versatility.
4Adaptability or versatility
If digital assistants translate and map commands across multiple languages, then language adaptability is improved, but computational resource consumption increases
Solution Approach 1:
Instead of creating and maintaining separate translation models for each language pair, the system uses a crowdsourced approach where users contribute translated action datasets. This copying strategy allows the system to leverage existing translations from the community rather than generating all translations computationally, significantly reducing resource consumption while maintaining broad language support.
Data Source
AI summary
Systems, methods, and non-transitory computer-readable storage media are provided for crowdsourcing actions and commands of a digital assistant application, irrespective of the languages spoken by users of the digital assistant application. Techniques described herein enable the on-boarding of actions datasets, which include defined commands and actions that result therefrom, from client devices to a remote server device. More specifically, the described techniques facilitate the proper on-boarding, distribution, and retrieval of action datasets regardless of the command language employed by users of the digital assistant application to invoke a properly-determined corresponding action.


