Machine learning in Education
summary
A crawler pulls YouTube videos with Spanish and English closed captions and a classifier initially labels 62 percent correctly, then an audio analyzer plus sliding-window feature extraction raises accuracy to 95 percent. The pipeline uses deep neural nets trained over a few hundred iterations, runs in-browser via web DNN / WebGL and web workers, and exposes classification as a service or client-side function; a live demo detects language reliably.
AI-written summary · about
chapters (4)
- 0:00Start
- 2:11How to improve learning foreign language with machine learning techniques
- 9:02Web workers Native performance in browser
- 17:50Visualisation Key to understanding
related talks
22mDeep learning module recognizes video sounds
52mMachine Learning Crash Course
45mTranscend Language Barriers with Custom Machine Learning APIs: um...
45mML: ML.NET: deep dive into dev AI
22mImplementation and analysis of deep learning models for speech classification
48mLearning Machine Learning
25mMachine learning algorithms as tools for student success prediction
28mHow to use Deep Learning to process audio data and classify sounds.