Voice Activity Detection in Noisy Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI systems face challenges in accurately recognizing user speech amidst ambient noise, leading to delayed or inappropriate responses due to misidentification of speech start and end points.
Innovation Solution
An electronic device equipped with a microphone, audio connector, and wireless communication circuit, utilizing a processor and memory to extract audio feature information from a first audio signal and recognize speech sections in subsequent audio signals, employing convolutional and recurrent neural networks to differentiate between voice and noise, ensuring accurate speech recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional voice activity detection is used in noisy environments, then the system can operate with simple processing, but speech recognition accuracy deteriorates due to ambient noise interference
Solution Approach 1:
The audio signal is divided into multiple frames, and each frame is processed independently to extract audio features. This segmentation allows the system to analyze speech and noise characteristics in small time windows, improving the accuracy of speech section detection in noisy environments by comparing features across multiple segmented frames rather than processing the entire audio signal as one unit.
2Speed
If the system responds quickly to detected speech, then response speed improves, but response accuracy deteriorates due to premature or incorrect speech start/end detection
Solution Approach 1:
The system uses audio feature information extracted from the first audio signal as feedback to improve speech section recognition in the second audio signal. By continuously analyzing audio features and using previous detections to inform current speech section identification, the system achieves more accurate start and end point detection while maintaining responsive performance.
Data Source
AI summary
According to various embodiments, an electronic device may include: a microphone; an audio connector; a wireless communication circuit; a processor operatively connected to the microphone, the audio connector, and the wireless communication circuit; and a memory operatively connected to the processor, wherein the memory may store instructions that, when executed, cause the processor to: receive a first audio signal through the microphone, the audio connector, or the wireless communication circuit, extract audio feature information from the first audio signal, and recognize a speech section in a second audio signal, received after the first audio signal through the microphone, the audio connector, or the wireless communication circuit, using the audio feature information.