Voice Dial Pronunciation Modeling for Accurate Name Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional speech recognition technology inaccurately recognizes the recipient's name during voice dialing, especially in vehicle environments, leading to user inconvenience.
Innovation Solution
A method and apparatus that generate multiple patterns based on the number and type of values in a recipient's name field, model a pronunciation dictionary, and prioritize candidates with a call history for accurate recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional speech recognition technology is used for voice dialing in vehicles, then the system is simple and easy to operate, but the recognition accuracy of recipient names is low
Solution Approach 1:
The system performs preliminary actions by generating multiple pronunciation patterns and modeling a pronunciation dictionary before speech recognition occurs. Call history data is pre-processed and stored, and pronunciation patterns are generated in advance based on name field values, enabling more accurate recognition during actual voice dialing without adding real-time complexity
Solution Approach 2:
The speech recognition system is segmented into multiple independent modules: call history analysis module, pronunciation pattern generation module, pronunciation dictionary modeling module, and speech recognition module. Each module handles a specific task, improving overall accuracy while maintaining manageable system complexity through functional decomposition
2Measurement precision
If multiple pronunciation patterns are generated and pronunciation dictionary is modeled to improve recognition accuracy, then recipient name recognition accuracy improves, but the processing time and computational complexity increase
Solution Approach 1:
Pronunciation patterns and the pronunciation dictionary are generated and modeled in advance during idle periods or when call history data is available, not during active speech recognition. This preliminary processing stores pre-computed patterns that enable fast lookup and matching during actual voice dialing, reducing real-time processing time while maintaining high accuracy
Solution Approach 2:
The system generates multiple pronunciation patterns (excessive action) for each name entry based on different interpretations of name field values. While this creates more patterns than strictly necessary, it comprehensively covers various pronunciation possibilities, ensuring high recognition accuracy. The system then uses efficient matching algorithms to quickly identify the correct pattern during recognition
Data Source
AI summary
A method and apparatus for providing voice dial. An aspect of the present disclosure provides a method for providing a voice dial, the method comprising: obtaining a phone book and a call history of a user, wherein the phone book and the call history include names and phone numbers of recipients; comparing a call history acquired in a present period with a call history acquired in a preceding period; determining whether a new call history exists, wherein the new call history is defined by a name or a phone number of a recipient exists in the call history acquired in the present period but does not exist in the call history acquired in the preceding period; checking a number of values included in a name field of the receipient in the phone book when the new call history exists; and modeling a pronunciation dictionary based on a new pattern combining the values based on a predetermined rule.