Reference Audio Extraction for Voice Recognition Echo Cancellation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current voice recognition systems face challenges in accurately processing spoken commands amidst external multimedia audio, leading to inoperativeness, false positives, and privacy issues due to the inability to compensate for external audio sources effectively.
Innovation Solution
A system that determines which controllable device an audible command is directed towards by using time and date stamps, amplitude comparison, and speech recognition algorithms, along with noise reduction processing and sensor integration to filter out noise energy and amplify command energy, thereby improving the accuracy of voice recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If voice recognition systems process audio signals in the presence of external multimedia audio, then the system can operate continuously and respond to commands, but the accuracy of voice recognition deteriorates due to noise interference and false positives
Solution Approach 1:
The patent extracts and removes external multimedia audio signals from the mixed audio input by comparing the audio signal from microphones with a reference signal from the external audio source. This extraction process isolates the voice command signal from the noise of external multimedia audio, enabling continuous operation while maintaining recognition accuracy.
Solution Approach 2:
The patent introduces an intermediary reference signal from the external multimedia audio source to mediate the separation of voice commands from background noise. By using this reference signal in correlation-based processing, the system can identify and remove external audio components while preserving the actual voice commands.
2Measurement precision
If the system uses multiple microphones and complex signal processing to improve voice recognition accuracy, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent replaces complex mechanical signal processing approaches with correlation-based digital signal processing. Instead of using multiple microphones with complex spatial processing, the system uses correlation between the microphone input and a reference signal to efficiently separate voice commands from external audio, reducing overall system complexity.
3Speed
If the system continuously monitors audio for commands, then response time improves, but false positives increase due to misinterpretation of noise
Solution Approach 1:
The patent implements feedback through correlation processing where the reference signal from external audio sources is continuously compared with the microphone input. This feedback mechanism allows the system to identify when audio patterns match external sources rather than actual commands, reducing false positives while maintaining continuous monitoring for rapid response.
Data Source
AI summary
Disclosed is a method for delaying audio from a received signal prior to processing in an acoustic echo cancellation device, comprising: receiving an audio signal at an audio extraction and delay device; determining a first amount of delay to apply to the received audio signal such that acoustic echo cancellation (AEC) processing can occur; applying the first amount of delay to the received audio signal; transmitting the un-delayed audio signal to an AEC circuit; and transmitting the delayed audio signal to one or more speakers.