This project was completed as a part of the Honors portion of the Sequence Models Course on Coursera.
Credit to DeepLearning.AI and the Coursera platform for providing the course materials and guidance.
In this assignment, my objective is to create a speech dataset and develop an algorithm for trigger word detection, also known as keyword detection or wake word detection. Trigger word detection technology enables devices like Amazon Alexa, Google Home, Apple Siri, and Baidu DuerOS to wake up when a specific word is spoken.
For this exercise, the trigger word will be "activate," and the algorithm will produce a "chiming" sound whenever it detects the word "activate" in an audio clip. By the end of this assignment, I aim to record a clip of myself speaking, and successfully have the algorithm trigger the chime upon detecting the word "activate."
Moreover, I plan to explore the possibility of extending this project to run on my laptop, allowing it to respond to the word "activate" by initiating specific actions, such as opening a favorite app, turning on a network-connected lamp, or triggering other events.
Throughout this assignment, I will gain valuable insights into structuring a speech recognition project, synthesizing and processing audio recordings to create training and development datasets, training a trigger word detection model, and making predictions. This will equip me with essential skills in developing speech-based applications and exploring the fascinating world of trigger word detection.
