Spotting and solving everyday problems with machine learning

on YouTube ↗13m

" summary

An explicit checklist of what gets built and what actually works: an AI-powered family video archive that uploads video to Cloud Storage, triggers the video intelligence API via Cloud Functions, stores transcripts and labels in BigQuery, and reliably finds concrete moments like a Barbie bike, birthday cakes (text read from icing) and first steps via fuzzy text matching; a semantic-search demo using the universal sentence encoder (TensorFlow JS) for headline matching; PDF→speech, auto-dub via Translation + Text-to-Speech, and pose tracking for tennis-serve angles.

AI-written summary · about

" chapters (8)

1. 0:00Start
2. 4:59Understanding Language
3. 5:30Embeddings
4. 6:27Semantic Reactor
5. 7:32Universal Sentence Encoder
6. 8:04Machine Learning To Convert between Different Data Types
7. 8:52Image to Text
8. 11:20Pose Tracking

" related talks

• 44mThe age of applied AIHenk Boelman · Build Stuff 2020
• 47mBuilding smarter apps with Machine Learning, from magic to realityLaurent Picard · GDG DevFest Poland 2018 in Wrocław
• 57mBuilding smarter solutions with Machine Learning, from magic to realityLaurent Picard · code.talks commerce 2019
• 54mBuilding smarter apps with Machine Learning, from magic to realityLaurent Picard · Voxxed Days Cluj-Napoca 2018
• 01h 00mBoost your App with Machine Learning APIsLaurent Picard · ML Conference 2018 Spring
• 36mLeveraging Machine Learning APIs to Write Better SoftwareSuz Hinton · Oredev 2017
• 34mI, HumanLéonie Watson · Beyond Tellerrand Berlin 2018
• 36mCloud AI Services: What they are and how to use themKarl Weinmeister · Global Artificial Intelligence Conference 2020