Voiceprint Identification for Voice Recognition Error Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional voice recognition systems in electronic devices suffer from high recognition error rates due to interference from external noise, mistakenly identifying sounds other than the user's voice input.
Innovation Solution
A method using voiceprint identification, where a received voice fragment is cut into sub-voice clips, and each clip is analyzed through a voiceprint identification flow using Mel-Frequency Cepstral Coefficients to determine appropriateness, with only matching clips being used for voice recognition, thereby reducing errors from noise interference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If voice recognition is performed on all received voice fragments, then processing speed is maintained, but recognition accuracy deteriorates due to noise interference
Solution Approach 1:
The patent segments the voice recognition process by dividing received voice fragments into multiple sub-voice clips and processing them separately through voiceprint identification. This segmentation allows the system to identify and filter out non-user voice segments (noise) while maintaining processing efficiency on valid segments only.
Solution Approach 2:
The patent applies preliminary voiceprint identification action before full voice recognition processing. By performing voiceprint matching on sub-voice clips first, the system pre-filters valid user voice segments, ensuring that subsequent voice recognition is only performed on authenticated segments, thereby improving accuracy without proportionally increasing overall complexity.
2Reliability
If voiceprint identification is applied to all sub-voice clips, then recognition accuracy is improved, but processing time increases
Solution Approach 1:
The patent applies partial action by performing voiceprint identification on individual sub-voice clips rather than on the entire voice fragment at once. This allows the system to process only the necessary portions of the audio data, applying voiceprint verification selectively to segments that need authentication, thereby reducing overall processing time while maintaining accuracy.
Solution Approach 2:
By dividing the voice fragment into multiple sub-voice clips and processing them in parallel or sequential batches, the patent reduces the time complexity of voiceprint identification. Each sub-voice clip is processed independently, allowing for more efficient utilization of processing resources and reducing the total time required compared to processing the entire fragment as a single unit.
Data Source
AI summary
A method for using voiceprint identification to operate voice recognition and electronic device thereof are provided. The method includes the following steps: receiving a specific voice fragment; cutting the received specific voice fragment into a plurality of specific sub-voice clips; performing a voiceprint identification flow to the specific sub-voice clips, respectively; determining whether each of the specific sub-voice clips is an appropriate sub-voice clip according to a result of the voiceprint identification flow; and capturing the appropriate sub-voice clips and operating a voice recognition thereto.