Facial Vibration Input for Smart Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing smart device input methods, such as typing on small screens or voice recognition, face challenges like difficulty in use with gloves or ambient noise, and are susceptible to replay and imitation attacks, necessitating a more accurate and efficient text input solution.
Innovation Solution
A smart device input method based on facial vibration signals, where a vibration sensor collects signals, extracts Mel-frequency cepstral coefficients, and uses a trained hidden Markov model to convert these signals into text input, avoiding noise and attack issues through real-time correction and self-adaptive mechanisms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If typing input through a virtual keyboard on a smart watch screen is used, then text input can be performed on smart devices, but it is difficult for users to perform typing input because the smart watch has a small screen
Solution Approach 1:
The patent replaces the mechanical typing action on a small keyboard with facial vibration detection. Instead of requiring finger movements on a limited screen area, the system captures vibrations from facial muscles during speech and converts them into text input, eliminating the need for a large physical keyboard while maintaining typing capability.
Solution Approach 2:
The patent introduces facial vibration signals as an intermediary between the user's intent to input text and the actual text output. Rather than directly mapping finger presses to characters, the system uses facial vibrations during speech as a mediator to capture input intent, enabling text entry without direct manual interaction with the small keyboard.
2Ease of operation
If handwriting input by finger tracking in the air is used, then users can input text without using hands for typing, but this input method is too slow and when users hold something in their hands, this handwriting input method is not applicable
Solution Approach 1:
The patent utilizes the periodic nature of facial vibrations that occur naturally during speech. Instead of requiring continuous or deliberate hand movements, the system captures rhythmic vibrations from facial muscles as the user speaks, converting these periodic physiological signals into text input at the natural pace of speech.
Solution Approach 2:
The patent enables the user's own facial physiology to serve as the input mechanism. During normal speech, the facial muscles naturally produce vibrations that the system captures and converts into text, eliminating the need for separate typing actions or conscious hand movements, thus achieving hand-free input at speech speed.
3Productivity
If traditional speech recognition technologies are used, then text input can be performed without typing, but these technologies are susceptible to ambient noise as well as replay attacks and imitation attacks
Solution Approach 1:
The patent extracts the vibration signal component from the overall speech signal. Instead of processing the complete acoustic speech signal that can be captured by microphones and subjected to noise and replay attacks, the system isolates and processes only the facial vibration signals, which are harder to replicate or spoof remotely.
Solution Approach 2:
The patent introduces facial vibration signals as a more secure intermediary for authentication and input verification. These vibrations serve as a biometric mediator that is harder to replay or imitate compared to acoustic speech signals, providing enhanced security while maintaining the convenience of speech-based input.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method provides efficient text input on small devices, resistant to ambient noise and attacks, with improved accuracy and robustness, enabling effective text entry even when hands are occupied.
Implementation Method 1
collecting a facial vibration signal generated when a user performs voice input
Data Source
AI summary
A smart device input method based on facial vibration includes: collecting a facial vibration signal generated when a user performs voice input; extracting a Mel-frequency cepstral coefficient from the facial vibration signal; and taking the Mel-frequency cepstral coefficient as an observation sequence to obtain text input corresponding to the facial vibration signal by using a trained hidden Markov model. The facial vibration signal is collected by a vibration sensor arranged on glasses. The vibration signal is processed by: amplifying the collected facial vibration signal; transmitting the amplified facial vibration signal to the smart device via a wireless module; and intercepting a section from the received facial vibration signal as an effective portion and extracting the Mel-frequency cepstral coefficient from the effective portion by the smart device.