Display Device Voice Recognition Acoustic Echo Cancellation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge is to accurately recognize a user's voice when an external speaker, such as a sound bar, is connected to a display device, as the user's voice is often mixed with the acoustic signal output from the speaker, and the installation position and output characteristics of the external speaker vary, making it difficult to remove the acoustic signal effectively.
Innovation Solution
A display device with a controller that measures the delay of the acoustic signal output from the external speaker using noise outside the audible frequency range, allowing for the removal of this signal from the microphone input, thereby isolating the user's voice. This involves using a noise other than the audible frequency to determine the delay and applying it to a cross-correlation function, and storing a gain table to correct for different output characteristics of external speakers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the acoustic signal from the external speaker is removed from the microphone signal, then the voice recognition accuracy is improved, but the complexity of signal processing increases
Solution Approach 1:
The system performs preliminary actions by measuring the delay between the external speaker output and microphone input in advance, and storing this delay information for later use in acoustic signal removal. This preliminary measurement simplifies the real-time processing by pre-calculating the delay parameters needed for acoustic echo cancellation.
Solution Approach 2:
The system uses feedback by continuously monitoring the microphone signal and comparing it with the expected acoustic signal from the external speaker. The delay information obtained from the feedback loop is used to adjust and refine the acoustic signal removal process, improving voice recognition accuracy while managing processing complexity.
2Measurement precision
If the delay measurement is performed using audible frequency noise, then the delay can be measured, but the user's viewing experience is disturbed
Solution Approach 1:
The system changes the frequency parameter of the measurement signal by using noise at non-audible frequencies (ultrasonic or infraasonic ranges) instead of audible frequencies. This parameter change allows accurate delay measurement while avoiding disturbance to the user's viewing experience, as the measurement noise falls outside the human hearing range.
3Measurement precision
If the acoustic signal removal is adapted to different external speaker characteristics, then the voice recognition accuracy is improved, but the system adaptability requirements increase
Solution Approach 1:
The system performs self-service by automatically measuring and storing delay information specific to each external speaker's installation conditions. The display device autonomously adapts to different speaker characteristics without requiring manual configuration or complex adaptation algorithms, simplifying the system's response to varying acoustic environments.
Solution Approach 2:
The system performs preliminary action by measuring and storing delay information for each external speaker before actual voice recognition operations. This pre-characterization of the acoustic path allows the system to handle different speaker configurations effectively, improving adaptability while maintaining processing efficiency.
Data Source
AI summary
A display device including an external device interface configured to connect the display device to an external speaker; a microphone configured to receive a microphone signal; and a controller configured to: transmit an inaudible test signal to the external speaker to measure a signal delay between the external speaker and the display device, and extract a voice signal by removing an acoustic signal output from the external speaker from the microphone signal based on the measured signal delay.