Multilingual Voice Query Search Using Profile-Based Language Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Voice recognition systems often fail to recognize words and phrases spoken in multiple languages or with accents, leading to inadequate search results.
Innovation Solution
A media guidance application that applies multiple language models to voice queries based on user profiles, automatically updating language settings to provide search results in multiple languages and ranking them based on phonetic matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single language model is used for voice recognition, then the device can process queries efficiently, but it cannot recognize words and phrases spoken in multiple languages or with accents
Solution Approach 1:
The system implements a universal voice recognition framework that can handle multiple languages and accents through a single unified interface. The media guidance application automatically selects and switches between different language models based on the detected query language, making the system multi-functional without requiring users to manually configure each language model.
Solution Approach 2:
The system dynamically adapts its language model selection based on the incoming query. Instead of using a static single language model, the system detects the language of the voice query and dynamically switches to the appropriate language model, allowing the recognition capability to be flexible and adaptive to different languages and accents.
2Adaptability or versatility
If multiple language models are applied to voice queries, then search results can be provided in multiple languages, but the processing time and computational resources increase
Solution Approach 1:
The system performs preliminary language detection on the voice query before executing the full search process. By identifying the query language in advance, the system can pre-select the appropriate language model and prepare relevant search parameters, avoiding the need to process all possible language models simultaneously and thus reducing overall processing time.
Solution Approach 2:
The search process is segmented into distinct stages: language detection, language model selection, and search execution. This segmentation allows the system to handle only the necessary language model for the detected query language, rather than processing multiple language models in full, thereby reducing computational overhead and processing time while still providing multi-language capability.
3Reliability
If voice recognition is limited to a single language, then the recognition accuracy is high, but users who speak multiple languages or have accents cannot be served
Solution Approach 1:
The system replaces the mechanical approach of using a single fixed language model with an intelligent system that automatically detects the query language and selects the appropriate language model. This substitution maintains high recognition accuracy for each specific language while providing adaptability to multiple languages and accents through automated language identification and model selection.
Data Source
AI summary
Systems and methods for a media guidance application that generates results in multiple languages for search queries. In particular, the media guidance application resolves multiple language barriers by taking automatic and manual user language settings and applying those settings to a variety of potential search results.