Building Qrio, a Bot That Plays Videos for My Toddler
summary
A home robot that recognizes toys (about 20 toy classes), identifies the person, and launches matching YouTube Kids videos is built and running on an NVIDIA Jetson Nano with a Raspberry Pi camera; object detection is a TensorRT-optimized model for real-time inference, speech uses Amazon Polly TTS, animation uses a game-engine skeleton system, and video playback is automated with Selenium. End-to-end latency lands under five seconds in demos and the system proves practical for an active toddler.
AI-written summary · about
chapters (5)
- 0:00Introduction to Qrio
- 14:00Building the video bot
- 23:00Text-to-speech integration
- 30:00Final demo and features
- 40:32Q&A and closing
related talks
17mRobot uses toddler-like self exploration for the development of body representations
21mAI, Embedded System and Robotics
13mSpotting and solving everyday problems with machine learning
43mBuilding the Duckling Droid: AI, Robotics, and Computer Vision in Action
34mBuilding Owly: An AI Comic Video Generator for My Son
28mBuild Your Child Their Very Own Node js Frozen Robot
44mAdventures in Puppy Training with A.I. and a Raspberry Pi
18mRobotics: why now?