WiMi Hologram Cloud Inc (WIMI)

[StockpickerCAPC]

WiMi Develops A Highly Reliable SSR System Based on Multi-feature Signal Perception.

Source
https://www.newstrail.com/wimi-develops-a-highly-reliable-ssr-system-based-on-multi-feature-signal-perception/

January 30, 2023

As time goes on, human-computer interaction is no longer limited to keyboard input and joystick operation but is taking on more innovative forms, such as finger movements, the vibration of sound waves, and the rotation of the eyeballs and tongue. These movements allow for the transfer of information and the “dialogue” between humans and machines. As technology continues to evolve, organic user interfaces:
- biometric sensors,
- skin displays, and
- even direct connections between the brain and computers, are beginning to emerge and will undoubtedly have a significant impact on human life.
This will undoubtedly have a substantial impact on human life. Recently, WiMi Hologram Cloud, Inc. (NASDAQ:WIMI) announced a highly reliable silent speech recognition system based on multi-featured signal sensing technology. Through silent reading or body movement recognition, the system can convert electrical signals from body movements or vocal cord movements into speech, allowing the system to decipher what a person is trying to say without the user having to give voice commands, thus enriching the human-computer interaction scenario of “machine understanding human language.”

Current voice recognition technology is a hands-free interface mode for VR applications, but it has several drawbacks, such as limited usability in noisy environments or public places and limited access for those who need to speak loudly and clearly. The system developed by WiMi enables SSR (silent speech recognition) by using either limb movements or fEMG (facial electromyography) in a holographic AR environment. Holographic EMG signal data or vocal fold vibration signal data presented by a person’s face or limbs are first captured, and then the two types of data captured are pre-processed separately. After feature extraction and fusion, deep learning is used to perform recognition and finally send the recognized command results to the receiving or controlled device.

To improve the system’s accuracy in classifying signals recorded at greater distances, WiMi developed a deep neural network-based classification method using an SSR system with fEMG in a holographic environment. The technique uses similar fEMG data previously collected from other objects, which is then transformed by holographic dynamic position distortion. When a person is speaking or thinking silently, the vocal-related limb muscles are influenced by the brain’s output nerves to produce different states of activity corresponding to the other content the brain wants to represent. The system acquires information from the brain by capturing surface EMG signals from the limb or vocal muscles and processing and recognizing them to perform voiced or unvoiced speech recognition. For feature recognition, the system uses independent CNN (Convolutional Neural Networks) to learn the features of each channel signal.
The developers have designed three structures:
- a one-dimensional convolutional network,
- a two-dimensional convolutional network, and
- a parallel convolutional network, configured the network model parameters, and optimized the network model structure.
For machine learning, the system uses Support Vector Machine, Random Forest, k-Nearest Neighbor, and Hidden Markov Model methods and continuously optimizes the model parameters during training and recognition. The model parameters are continuously optimized during training and recognition.

WiMi’s highly reliable SSR system based on multi-feature signal perception technology achieves silent speech recognition by combining holographic visual information and facial or limb electromyographic information to promote further development of speech recognition technology and expects to provide new ideas and methods for the current speech recognition field.