Voice Recognition Timing Synchronization for Noise Cancellation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Voice recognition systems face degradation due to unwanted audio interference from nearby sources, such as television speakers, which existing methods like directional microphones and acoustic beamforming struggle to effectively filter out, especially when the interference and desired signals come from similar directions or have variable timing due to unreliable WiFi connections.
Innovation Solution
The system employs an always-listening device with a microphone that adjusts the audio timing to match the timing of data received via a more deterministic RF4CE connection, allowing for effective filtering of unwanted audio interference by synchronizing audio data with RF4CE timing information, thereby isolating and removing interference from the desired voice commands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If directional microphones or acoustic beamforming are used to filter audio interference, then voice recognition accuracy may be improved, but they fail to effectively filter interference when it comes from similar directions or has variable timing
Solution Approach 1:
The patent introduces an intermediary approach by using a reference audio signal from the television audio output to identify and cancel interference. Instead of relying solely on spatial filtering (directional microphones) or beamforming, the system uses the known reference signal to create an interference model that can be subtracted from the microphone input, effectively handling interference from any direction and with variable timing.
Solution Approach 2:
The system dynamically adjusts timing parameters by introducing variable delays to synchronize the reference audio signal with the microphone captured signal. This timing adjustment allows the system to compensate for variable propagation delays and maintain effective interference cancellation even when interference timing varies, something that fixed directional patterns cannot handle.
2Ease of operation
If WiFi is used for audio data transmission, then device connectivity is improved, but timing reliability deteriorates due to connection quality variations
Solution Approach 1:
The patent implements dynamic timing adjustment by introducing variable delays that adapt to the actual timing of received audio packets. Instead of relying on fixed timing assumptions, the system continuously adjusts the reference signal timing to match the actual arrival times, compensating for WiFi jitter and connection variations. This dynamic approach maintains timing reliability while preserving WiFi connectivity benefits.
Solution Approach 2:
The system uses feedback from the actual packet arrival times to adjust timing delays. By monitoring when audio packets actually arrive and comparing this to expected timing, the system dynamically adjusts delay parameters to maintain synchronization, creating a closed-loop timing correction mechanism that compensates for WiFi reliability issues.
3Productivity
If audio from television speakers is captured by the microphone, then the device can process local audio, but voice command recognition is degraded by unwanted output audio
Solution Approach 1:
The patent extracts the harmful interference component by using the television audio output as a reference signal. The system identifies the interference portion in the microphone captured audio by correlating it with the reference signal, then removes this extracted interference component, leaving only the desired voice command signal for processing.
Solution Approach 2:
The system converts the harmful television audio output into a beneficial reference signal. By using the known television audio output as a reference, the system can actively identify and cancel this same signal when it appears as interference in the microphone input, transforming the harmful interference source into a useful tool for its own cancellation.
Data Source
AI summary
Systems, devices, and methods are described for reducing degradation of a voice recognition input. An always listening device may always be listening for voice commands via a microphone and may experience interference from unwanted audio such as from the output audio of television speakers. The always listening device may receive data associated with the output audio over a first communications channel. The always listening device may also receive, on a second communications channel, timing information associated with data. The always listening device may adjust admission of the audio received by the microphone to enable it to arrive at approximately the same time as the data received via the first communications channel. The unwanted output audio included in the audio received via the microphone may then be determined and may be removed so that a voice command in the audio received by the microphone may be processed.