Universal Voice AI Invocation Name Mapping for Broadcast Content

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The convenience of voice AI assistance services is compromised when users need to remember and switch between different invocation names for various broadcast stations and programs, leading to a less user-friendly experience.

Innovation Solution

A system that allows users to interact with voice AI services using a universal invocation name, which is replaced by a specific operational invocation name associated with each broadcast station or program, either locally or through a cloud-based alias skill, enabling seamless interaction across different content types.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If different invocation names are used for each broadcast station or program, then the voice AI service can be customized for specific content, but the user must remember and switch between multiple invocation names, reducing convenience

Engineering Contradiction:
Improvecustomization for specific contentVSAvoiduser convenience
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The patent implements a universal invocation name that can trigger multiple different skills or programs depending on the context. Instead of requiring separate invocation names for each broadcast station or program, a single universal name (e.g., 'AI assistant') is used, and the system automatically determines which specific skill to activate based on the current content being viewed. This resolves the contradiction by maintaining customization capability while eliminating the need for users to remember multiple invocation names.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent introduces a mediation layer between the user's voice input and the specific skills/programs. This intermediary system analyzes the current broadcast content, the user's utterance, and contextual information to automatically route the voice command to the appropriate skill. The mediator handles the complexity of skill selection internally, allowing users to interact with all content through a single universal invocation name without needing to know which specific skill will be activated.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If a single universal invocation name is used, then user convenience is improved, but the system must manage multiple skills and their corresponding specific invocation names internally

Engineering Contradiction:
Improveuser convenienceVSAvoidskill management complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent implements a feedback mechanism where the system continuously monitors the current broadcast content and provides this information to the skill selection logic. When a user invokes the universal name, the system receives feedback about what content is currently being viewed and uses this feedback to automatically select the appropriate skill. This feedback loop allows the system to manage multiple skills internally while presenting a simple universal interface to users, as the skill selection is dynamically determined based on real-time content feedback.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent performs preliminary actions by pre-registering and storing the relationships between broadcast content and corresponding skills in advance. The system maintains a database or configuration that maps different broadcast stations, programs, and content types to their associated skills. This preliminary organization of skill information allows the system to quickly and automatically resolve which skill to activate when a universal invocation name is used, without requiring complex real-time decision-making or increasing operational complexity for the user.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3780641B1Information processing device, information processing method, transmission device and transmission method
Publication Date: 2023.01.25 SONY GROUP CORP
  • EP3780641B1 patent drawingFigure 1
  • EP3780641B1 patent drawingFigure 2
  • EP3780641B1 patent drawingFigure 3

AI summary

The present technology relates to an information processing apparatus, information processing method, transmission apparatus, and transmission method, capable of improving the convenience of a voice AI assistance service used in cooperation with content. Provided is an information processing apparatus including a processing unit configured to process, in using a voice AI assistance service in cooperation with content, specific information associated with a universal invoking name included in a voice uttered by a viewer watching the content on the basis of the universal invoking name and association information, the universal invoking name being common to a plurality of programs that perform processing corresponding to the voice uttered by the viewer as an invoking name used for invoking the program, the association information being associated with the specific information to each of the programs. The present technology can be applied to a system in cooperation with a voice AI assistance service, for example.