Google tool teaches machine learning to six-year-olds

Teachable Machine 2.0 is an unintimidating no-code platform that eases kids into understanding machine learning

Google HQ

Google has launched the second iteration of its no-code Teachable Machine so that inexperienced users can take their bespoke machine learning (ML) models and apply them to projects such as classroom activities.

Teachable Machine 2.0 carries over the features from the original, allowing users to record images and video from a webcam and use them to train ML models for tasks like pattern recognition. Now, these same models can be taken and exported to websites, apps and physical machines.

Open source curriculums are making use of the tool to give children their first taste of ML, without the intimidating aspect of learning to code.

One such example is a programme run out of MIT's Media Lab by education researcher Blakeley H. Payne for six to 10-year olds. The children are invited to the lab where they use Teachable Machine 2.0, among other things, to get a wider knowledge base about technology and what it can do.

"Parents - especially of girls - often tell me their child is nervous to learn about AI because they have never coded before," said Payne. "I love using Teachable Machine in the classroom because it empowers these students to be designers of technology without the fear of 'I've never done this before.'"

The tool is entirely browser-based and all the data that's fed into it stays on the origin computer while the processing is done in the browser.

Teachable Machine can record from a computer's webcam and microphone and be trained to recognise images, sounds or poses. It can identify different people or objects and detect when they leave or return to the shot.

Other real-world use cases involve helping those with impaired speech use voice-powered computer products. For example, neurologic conditions that impair speech, such as motor neurone disease, can impede an individual's ability to interact with software. But, Teachable Machine can take audio, turn it into a spectrogram and be trained to correct the sound of speech that isn't produced normally.

Elsewhere, educators at New York University's interactive telecommunications program used the tool to create video games where characters could be controlled using hand gestures, thanks to the pose recognition feature.

You can get started with Teachable Machine too by following this link and basic tutorials can found on Google's page.

Featured Resources

The complete guide to changing your phone system provider

Optimise your phone system for better business results

Download now

Simplify cluster security at scale

Centralised secrets management across hybrid, multi-cloud environments

Download now

The endpoint as a key element of your security infrastructure

Threats to endpoints in a world of remote working

Download now

2021 state of IT asset management report

The role of IT asset management for maximising technology investments

Download now

Recommended

Workday releases new Accounting Center to help manage financial data
chief financial officer (CFO)

Workday releases new Accounting Center to help manage financial data

30 Oct 2020
MarqVision detects counterfeit products with deep learning and AI
intellectual property

MarqVision detects counterfeit products with deep learning and AI

18 Sep 2020
The IT Pro Podcast: Attack of the AI hackers
artificial intelligence (AI)

The IT Pro Podcast: Attack of the AI hackers

14 Aug 2020
MIT develops AI tech to edit outdated Wikipedia articles
artificial intelligence (AI)

MIT develops AI tech to edit outdated Wikipedia articles

13 Feb 2020

Most Popular

Do smart devices make us less intelligent?
artificial intelligence (AI)

Do smart devices make us less intelligent?

19 Oct 2020
Best MDM solutions 2020
mobile device management (MDM)

Best MDM solutions 2020

21 Oct 2020
What is Neuralink?
Technology

What is Neuralink?

24 Oct 2020