Concurrent Voice Assistants With Hotword Audio Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automated assistants require manual configuration changes to switch between different hotwords, leading to inefficient resource utilization and negative user experiences when unsupported tasks are attempted.
Innovation Solution
Implementing multiple concurrent voice assistants that listen for different hotwords, allowing users to invoke specific assistants without configuration changes, and managing microphone and speaker access to prevent interference between active sessions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a single automated assistant is selected and configured, then the assistant can handle tasks efficiently, but the user cannot switch between different assistants without manual configuration changes
Solution Approach 1:
The system segments the automated assistant functionality into multiple independent assistant instances, each with its own hotword detection and task handling capabilities. This allows users to switch between assistants by speaking different hotwords without manual configuration, as each assistant operates independently with its own activation trigger.
Solution Approach 2:
The system implements a universal hotword detection mechanism that can identify multiple different hotwords (e.g., 'Hey Assistant', 'OK Assistant') and route them to different assistant instances. This multi-functional approach allows a single system to support multiple assistants simultaneously, eliminating the need for manual configuration switching.
2Adaptability or versatility
If multiple automated assistants are supported concurrently, then users can invoke specific assistants without configuration changes, but resource utilization becomes inefficient when unsupported tasks are attempted
Solution Approach 1:
The system performs preliminary hotword detection and classification before fully activating an assistant. When a hotword is detected, the system first identifies which assistant instance should handle the request, then selectively activates only that specific assistant. This preliminary routing action prevents computational resources from being wasted by having all assistants continuously monitor and process all audio data.
Solution Approach 2:
Each assistant instance maintains its own localized hotword detection and processing capabilities, but the system implements selective activation where only the relevant assistant processes audio data after hotword identification. This local quality approach allows each assistant to be optimized for its specific domain while the system manages resource allocation efficiently by activating only the necessary assistant instance.
3Productivity
If multiple automated assistants process audio data simultaneously, then comprehensive task handling is achieved, but audio data from one assistant may be processed by another assistant causing interference
Solution Approach 1:
The system introduces an intermediary audio data stream management mechanism that routes audio data to specific assistant instances based on the detected hotword. The intermediary layer (hotword detection and routing system) ensures that audio data is directed only to the appropriate assistant instance, preventing cross-assistant interference while maintaining comprehensive task handling capabilities.
Solution Approach 2:
The system dynamically assigns audio data streams to specific assistant instances based on real-time hotword detection. When a hotword is detected, the audio processing dynamic routing mechanism activates only the relevant assistant instance for that specific audio stream. This dynamic approach allows multiple assistants to operate simultaneously with comprehensive task handling while preventing interference through selective activation and data routing.
Data Source
AI summary
Techniques are described herein for concurrent voice assistants. A method includes: providing first and second automated assistants with access to one or more microphones; receiving, from the first automated assistant, an indication that the first automated assistant has initiated a first session, and in response: continuing providing, to the first automated assistant, access to the one or more microphones; discontinuing providing, to the second automated assistant, access to the one or more microphones; and preventing the second automated assistant from accessing one or more portions of an output audio data stream; receiving, from the first automated assistant, an indication that the first session has ended, and in response: continuing providing, to the first automated assistant, access to the one or more microphones; resuming providing, to the second automated assistant, access to the one or more microphones; and resuming providing, to the second automated assistant, the output audio data stream.