AI Speech Recognition with Multi-Language Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current speech recognition devices struggle with recognizing and understanding speech commands that contain multiple languages, leading to low recognition rates and difficulty in grasping the intent behind such commands.
Innovation Solution
An AI device and method that separates speech data by language, performs natural language understanding processing for each language, and integrates the results to recognize user intent and generate appropriate responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a speech recognition model classifies and processes speech according to each language separately, then the recognition accuracy for single-language speech is improved, but the recognition rate for mixed-language speech deteriorates
Solution Approach 1:
The patent segments the speech processing task by separating text data into multiple language-specific data pieces using language identification and separation modules. Each language segment is then processed by its dedicated natural language understanding model, enabling accurate recognition of mixed-language speech while maintaining the benefits of language-specific processing
Solution Approach 2:
The patent creates a universal speech recognition system that can handle both single-language and mixed-language inputs. The system integrates multiple language-specific NLU models into a unified framework that automatically detects and processes different language combinations, making the system adaptable to various speech scenarios
2Ease of operation
If speech is processed as a whole without language separation, then the processing simplicity is maintained, but the ability to grasp intent in multi-language commands deteriorates
Solution Approach 1:
The system automatically segments mixed-language speech into language-specific components without requiring manual intervention. The language separation module divides the input text into distinct language segments, which are then processed by appropriate NLU models to accurately capture intent while maintaining operational simplicity for the user
Solution Approach 2:
The patent introduces language identification and separation modules as intermediary components between speech input and intent analysis. These intermediaries automatically detect languages, separate text segments, and route them to appropriate processing models, preserving intent understanding accuracy while keeping the overall system operation simple and transparent
Data Source
AI summary
An artificial intelligence device includes a microphone configured to acquire speech including a plurality of languages, and a processor configured to generate, from the speech, text data corresponding to the speech, generate a plurality of pieces of separated data acquired by separating the text data for each language, perform natural language understanding processing corresponding to a language of each of the plurality of pieces of separated data to generate a natural language understanding processing result for each of the plurality of pieces of separated data, acquire command information about a command to be instructed by the speech and slot information about an entity subjected to the command, based on the natural language understanding processing result, perform an operation corresponding to the speech based on the command information and the slot information, and generate a response based on a result of performing the operation.