Скачать или смотреть Final Steps & Series Wrap-Up | TinyML Wake-Word Detection

In this final video of the TinyML Wake-Word Detection Series, we bring everything together and close out the project. We explore how to record your own audio, apply data augmentation, and design custom models — before wrapping up the series.

👉 What we cover in this session:

Recording custom audio for training datasets

Using data augmentation (noise, pitch variation, shifting) to strengthen models

Exploring custom architectures for extended functionality

Wrapping up the wake-word detection workflow and summarizing key takeaways

This session serves as the conclusion of the Wake-Word Detection series, setting the stage for the next TinyML applications.

📚 Reference material:

Wes McKinney, Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython, Second Edition, O’Reilly Media, 2018.

Chandra S.S., Vinod, Hareendran S., Anand, Machine Learning: A Practitioner’s Approach, PHI Learning Pvt. Ltd., 2021.

Pete Warden, Daniel Situnayake, TinyML: Machine Learning with TensorFlow Lite on Arduino and Ultra-Low-Power Microcontrollers, O’Reilly Media, 2019.

#TinyML #WakeWordDetection #CustomModels #DataAugmentation #TensorFlowLite #ArduinoNano33BLE #EmbeddedAI #EdgeAI #AIoT