Intelligent voice control, UV, disinfection lamp solution, offline Chinese and English speech recognition control module
The scheme adopts high-sensitivity analog microphone recording, and uses steady-state and dynamic noise filtering algorithms to dynamically adjust the recording volume. After recording noise reduction, the main material data with high signal-to-noise ratio is sent to the recognition engine for recognition through speech enhancement, which ensures the high accuracy and anti-noise ability of different distance recognition, and is suitable for speech recognition in real and complex scenes in long-distance or noisy environment Voice control. It can be controlled by 5-10m far-field speech recognition. In this scheme, we can set the delay time of UV disinfection lamp to reduce unnecessary UV radiation; By means of delayed light on and voice warning, users are reminded to leave the site with UV radiation risk quickly after turning on the light; The working time of the disinfection lamp can be determined according to the customer's requirements, 15 minutes, 30 minutes, 60 minutes... The UV disinfection lamp manufacturer only needs to add a microphone pickup hole and horn hole on the shell of the existing product, which can be applied, and quickly realize the intelligent upgrading of the ordinary UV disinfection lamp at low cost!
Technical characteristics
1. Adjustable identification distance: support 5-10m identification distance;
2. Background noise suppression: steady state and dynamic noise, ambient noise suppression > 20dB;
3. Local speech recognition (based on the new TDNN deep neural network recognition algorithm) supports pure off-line recognition;
4. The comprehensive recognition rate can reach more than 95%;
5. Support the customization of hundreds of local instructions, completely offline, no network, and the recognition time is less than 100 ms;
6. The misjudgment rate is very low.