Speech Recognition Platform Intent Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently interpreting and responding to voice commands in a home environment, particularly in determining the intent behind user speech and performing corresponding tasks.
Innovation Solution
A speech recognition platform that performs automatic speech recognition (ASR) on audio signals to identify domains and intents, and subsequently performs tasks such as playing music, purchasing items, or setting reminders, by interacting with remote computing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic speech recognition and intent identification are implemented, then user interaction efficiency is improved, but system complexity increases
Solution Approach 1:
The speech recognition platform is divided into distinct functional modules: audio signal reception, automatic speech recognition (ASR), domain identification, intent identification, and task execution. Each module handles a specific aspect of the processing pipeline, allowing for independent optimization and maintenance while collectively improving user interaction efficiency through automated interpretation and response to voice commands
Solution Approach 2:
The patent introduces intermediate processing layers between raw speech input and task execution. The domain identification module acts as an intermediary that categorizes speech into specific domains (e.g., music, news, weather) before intent analysis. This intermediary structure simplifies the overall system architecture by breaking down the complex speech-to-task pipeline into manageable stages with clear interfaces between them
2Adaptability or versatility
If multiple domains and intents are identified, then system versatility is improved, but processing time increases
Solution Approach 1:
The system performs preliminary domain identification before conducting detailed intent analysis. By first categorizing the speech into a specific domain (e.g., determining it's a music-related query), the system can then focus intent identification resources on domain-specific patterns and vocabularies. This preliminary action reduces the search space for subsequent intent analysis, maintaining versatility across multiple domains while reducing overall processing time through staged analysis
Data Source
AI summary
A speech recognition platform configured to receive an audio signal that includes speech from a user and perform automatic speech recognition (ASR) on the audio signal to identify ASR results. The platform may identify: (i) a domain of a voice command within the speech based on the ASR results and based on context information associated with the speech or the user, and (ii) an intent of the voice command. In response to identifying the intent, the platform may perform multiple actions corresponding to this intent. The platform may select a target action to perform, and may engage in a back-and-forth dialog to obtain information for completing the target action. The action may include streaming audio to the device, setting a reminder for the user, purchasing an item on behalf of the user, making a reservation for the user or launching an application for the user.