We do have lots of musical books, where you push the spot on the page and usually an animal plays some instrument.
At the moment, we feel he is a little too small at 2 for his own instruments, as they will be destroyed in pretty short order. They're high on the list of things to obtain when he's just a little more careful with things.
Drums can take a fair amount of abuse (within reason - better than stringed instruments, for example)... though you may not enjoy his early jam sessions. I gave my son a set of tuned handbells when he was 2, they were a hit, and hard to break.
How about taking into account how consumption of this content has changed? This kind of stuff is great for mobile consumption, which, as far as I understand, is growing at a rapid pace.
> This kind of stuff is great for mobile consumption
To whom? Most people I know are more incensed at autoplaying video on their smartphones even more than on the desktop. Their audio is more likely to be turned up on their phone, and it's harder to turn it off, which is what most people immediately do.
Sure. I don't have anything to link on the spot but this was/is/has been foreseeable for some time. Although it's all very cool and shiny - most business applications of machine learning remain squarely in the territory of classic algos like GLM & forests (random, boosted trees etc. etc.). As a fun note, advances like these highlight that data scientists etc. will not be beaten by more complex automated methods, but simply by speed. Much like the filing system that 'runs' whatever you're using to see these words (https://www.youtube.com/watch?v=EKWGGDXe5MA).
Edit: to elaborate... single model training runs are possible to do quite fast now, but knowing how to tune hyper parameters remains the 'voodoo' of the field. But the best hyper params are also possible to discover through brute force: try every combination you can! Today, you can use various heuristics to improve this process, but either way, being able to train whatever X times faster just means we can search hyper parameter space that much faster. The robots are coming :)
They could run 10x more experiments for the same cost and experiment with many more configurations, but soon enough there will probably be an algorithm that can do the same on an single computer. I am waiting for the moment neural networks will become as good as people in designing neural networks.
I agree there are examples that show what you suggest. There are far more examples to suggest the opposite though. If the technology becomes widely available and there are enough independent companies to promote competition rather than collusion, the prices have no choice but to go down.
