%0 Journal Article %1 noauthororeditor %A V, Viswanatha %A A.C, Ramachandra %A R, Raghavendra Prasanna %A Kakarla, Prem Chowdary %A PJ, Viveka Simha %A Mohan, Nishant %D 2022 %J Journal of Xi'an University of Architecture & Technology %K %N 7 %P 160-169 %R https://doi.org/10.37896/JXAT14.07/315417 %T IMPLEMENTATION OF TINY MACHINE LEARNING MODELS ON ARDUINO 33 – BLE FOR GESTURE AND SPEECH RECOGNITION %U https://app.box.com/s/hqjfpzawzddqylztyd6xlc5d7yuyg2gx %V 14 %X In this articlegesture recognition and speech recognitionapplications are implemented on embedded systems with Tiny Machine Learning (TinyML).The main benefit of using TinyML is its portability. It can be run on cheap microcontrollers with tiny batteries and low power consumption, and itcan easily integratemachine learning withvirtually anything. It also has the benefit of increased security due to local nature of computing.The benefit of using Arduino Nano 33 BLE sense is that it has a set of sensors embedded on the top, which gives us a lot of options to try ideas without having to generate the circuit to such sensors in prototyping board. It features 3-axis accelerometer, 3-axis gyroscope and 3-axis magnetometer. The gesture recognition,provides an innovative approach nonverbal communication. It has wide applications in human-computer interaction and sign language. Here in the implementationof handgesture recognition,TinyMLmodel istrained and deployedfromEdgeImpulseframeworkfor hand gesture recognitionand based on the hand movements,ArduinoNano 33 BLE device having 6-axis IMUcan find out the direction of movement of hand. The Speech is a mode of communication. Speech recognition is a way by which the statements or commands of human speech is understood by the computer which reacts accordingly. The main aim of speech recognition is to achieve communication between man and machine. Here in the implementationofspeech recognition,TinyMLmodel istrained and deployedfromEdgeImpulseframeworkfor speechrecognitionand based on thekeyword pronounced by human,ArduinoNano 33 BLE device having built-in microphonecan make an RGB LED glowlike red, green or blue based on keyword pronounced. The results of each applicationare obtained and listed in the results section and given the analysis upon the results.

@article{noauthororeditor, abstract = {In this articlegesture recognition and speech recognitionapplications are implemented on embedded systems with Tiny Machine Learning (TinyML).The main benefit of using TinyML is its portability. It can be run on cheap microcontrollers with tiny batteries and low power consumption, and itcan easily integratemachine learning withvirtually anything. It also has the benefit of increased security due to local nature of computing.The benefit of using Arduino Nano 33 BLE sense is that it has a set of sensors embedded on the top, which gives us a lot of options to try ideas without having to generate the circuit to such sensors in prototyping board. It features 3-axis accelerometer, 3-axis gyroscope and 3-axis magnetometer. The gesture recognition,provides an innovative approach nonverbal communication. It has wide applications in human-computer interaction and sign language. Here in the implementationof handgesture recognition,TinyMLmodel istrained and deployedfromEdgeImpulseframeworkfor hand gesture recognitionand based on the hand movements,ArduinoNano 33 BLE device having 6-axis IMUcan find out the direction of movement of hand. The Speech is a mode of communication. Speech recognition is a way by which the statements or commands of human speech is understood by the computer which reacts accordingly. The main aim of speech recognition is to achieve communication between man and machine. Here in the implementationofspeech recognition,TinyMLmodel istrained and deployedfromEdgeImpulseframeworkfor speechrecognitionand based on thekeyword pronounced by human,ArduinoNano 33 BLE device having built-in microphonecan make an RGB LED glowlike red, green or blue based on keyword pronounced. The results of each applicationare obtained and listed in the results section and given the analysis upon the results.}, added-at = {2022-07-27T09:02:32.000+0200}, author = {V, Viswanatha and A.C, Ramachandra and R, Raghavendra Prasanna and Kakarla, Prem Chowdary and PJ, Viveka Simha and Mohan, Nishant}, biburl = {https://www.bibsonomy.org/bibtex/2c96279fec444a9c93b94b4594af83f99/viswas123}, doi = {https://doi.org/10.37896/JXAT14.07/315417}, interhash = {75fa8b1b18d5b0787ad0cf82c5b59fdd}, intrahash = {c96279fec444a9c93b94b4594af83f99}, issn = {1006-7930}, journal = {Journal of Xi'an University of Architecture & Technology}, keywords = {}, language = {English}, month = {july}, number = 7, pages = {160-169}, timestamp = {2022-07-27T09:02:32.000+0200}, title = {IMPLEMENTATION OF TINY MACHINE LEARNING MODELS ON ARDUINO 33 – BLE FOR GESTURE AND SPEECH RECOGNITION}, url = {https://app.box.com/s/hqjfpzawzddqylztyd6xlc5d7yuyg2gx}, volume = 14, year = 2022 }