Speech Recognition User Identification via Parameter Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current speech recognition technologies fail to accurately determine the user issuing a speech instruction, leading to decreased accuracy and potential security risks, as they only recognize the content of the speech signal without identifying the user, allowing unauthorized access.
Innovation Solution
A method and device for speech recognition that performs analog to digital conversion of speech signals, extracts speech parameters such as fundamental frequency, sound pressure, and spectrum information, and matches these parameters with prestored values in a sample library to determine the user and execute control signaling only if the user has permission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If speech recognition only recognizes content without identifying user, then recognition speed is fast, but security and accuracy deteriorate
Solution Approach 1:
The speech recognition system is segmented into two independent modules: content recognition module and speaker identification module. The content recognition module processes the semantic meaning of speech, while the speaker identification module extracts and compares speech parameters (fundamental frequency, formant frequencies, spectral characteristics) from the same speech signal to identify the user. This segmentation allows the system to enhance security through user identification without significantly increasing overall system complexity, as both modules process the same input signal independently.
2Measurement precision
If speech parameters are extracted and matched with prestored values, then user identification accuracy is improved, but processing time increases
Solution Approach 1:
Speech parameters of multiple users are pre-recorded and stored in a database during system initialization or user registration phase. When speech recognition is performed, the system directly compares the extracted speech parameters (fundamental frequency, formant frequencies, spectral features) against these pre-stored parameter sets. This preliminary preparation of reference data eliminates the need for complex real-time analysis of user identity, significantly reducing processing time while maintaining high identification accuracy through parameter matching.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances the accuracy of speech recognition by identifying the user issuing the speech signal and ensures secure execution of control commands, preventing unauthorized operations by matching the speech parameters with prestored values in the sample library.
Implementation Method 1
performing analog to digital conversion of the first speech signal to generate a first digital signal after the analog to digital conversion
Data Source
AI summary
Embodiments of the present disclosure provide a method and device for speech recognition. The solution comprises: receiving a first speech signal issued by a user; performing analog to digital conversion on the first speech signal to generate a first digital signal after the analog to digital conversion; extracting a first speech parameter from the first digital signal, the first speech parameter describing a speech feature of the first speech signal; if the first speech parameter coincides with a first prestored speech parameter in a sample library, executing control signalling instructed by the first digital signal, the sample library prestoring prestored speech parameters of N users, N≧1. The solution can be applied in a speech recognition process and can improve the accuracy of speech recognition.