Explore speech-to-intent model training and deployment on microcontrollers in this 30-minute video tutorial. Learn an efficient approach for device control using speech recognition, bypassing traditional text transcription methods. Discover techniques for training domain-specific speech-to-intent models and deploying them on resource-constrained Cortex M4F-based development boards like the Wio Terminal. Follow along as the instructor demonstrates data processing, model training, and MCU inference code implementation. Gain insights into testing the inference on-device and potential improvements for speech recognition on microcontrollers. Access additional resources, including GitHub repositories and related TinyML talks, to further enhance your understanding of speech-to-intent technology for low-power, low-footprint devices.
Overview
Syllabus
Demo
Intro
What is Speech to Intent
Training code for reference model
Fluent.ai Speech commands dataset
Data processing and model training
MCU Inference code explanation
Testing the inference on device
Improvements and conclusion
Taught by
Hardware.ai
Related Courses
-
Deploying TinyML
-
TinyML Talks - Speech-to-Intent Model Deployment to Low-Power Low-Footprint Devices
-
Introduction to TinyML on Wio Terminal - Machine Learning on Microcontrollers
-
Wio Terminal TinyML Course - Audio Scene Recognition on Microcontrollers with Arduino IDE
-
Speech Recognition on Low Power Devices
-
Wio Terminal TinyML Course - People Counting and Azure IoT Central Integration - Part 3
