[CuriouslyC]

The way I see it, the biggest opportunities for the foreseeable future are VR and tools that leverage deep learning to enable unskilled people to produce high quality creative works (e.g. neural drawing, neural music making, neural writing, etc). Doing a side project in either of these areas is more likely to lead to something revolutionary, and even if you fail your job prospects will be greatly improved. On the other hand, both of these areas are hard, but in life, nothing ventured, nothing gained.

[fspear]

I agree however, creating something revolutionary in these spaces takes more than just fiddling with a couple of tensorflow tutorials or udacity courses.

For instance I've always wanted to have a music recommendation engine/app that instead of using collaborative filtering uses some kind of machine learning or deep learning algorithm(s) to identify lyric sentiment, complexity and melodic/rhythmic patterns to recommend me music based on that (across different genres). It turns out that something that seemed straightforward (I really thought I could tackle this after playing around with tensorflow and machine learning tutorials & MOCS) actually requires a lot of specialized/intimate knowledge about how these algorithms work and the theory behind them, I would need to spend quite some time studying (and understanding) before I can even begin coding this as a side project for instance.

[mindcrime]

Offhand, I would have thought that would be limited more to knowledge about audio signal processing and what-not, than the actual machine learning itself. I mean, yeah, you can get REALLY deep in the weeds with neural networks, but you can do an awful lot without doing so. If you ever feel like writing about this, I think a lot of people might find value in a blog post from you about what challenges you ran into and what you found out so far. I know I would.

[fspear]

Yes you are absolutely right, DSP knowledge is also a must. I didn't get that far, I started by identifying what makes people like a certain song and got it down to: Lyrical themes (and overall artist thematic) and melodic and rhythmic patterns. For instance by analyzing data from last.fm I found that a lot of people that listen to Black Metal also seem to have an interest in european/celtic/gaelic folk music, but not all only certain artists with specific lyrical themes and what seemed to me as specific melodic patterns.

Anyway the thing that stumped me was identifying/creating the "features" for rhythmic & melodic patterns, eventually I realized that I was way out of my league and had to drop it.

Later on I found that someone else was already working on this: http://benanne.github.io/2014/08/05/spotify-cnns.html

[0xdeaddead]

This is one of the hype filled and most valley answers I've ever read. You even used the word "revolutionary". Jony is that you?

[CuriouslyC]

What were you trying to accomplish with this comment?

[0xdeaddead]

Put some reality into the conversation.

The OP's opening sentence asked for a project to make $500-$1000, not ride a hype wave.

[CuriouslyC]

Do you think that is a good reason to be snarky?

The title of the OP's post wasn't "I need a way to make some quick extra cash". Most people I know work on side projects more for learning and career development. I gave an answer based on that prior belief.

Sometimes, hype exists because things are genuinely cool. I feel this is the case for VR and smart tools. On the other hand, I'd be inclined to agree with you about chat bots, for instance.

[mrfusion]

What do you recommend for a VR project?

[CuriouslyC]

A game would be the easiest thing to get started with. I think down the line, virtually exploring places via scene reconstruction from multiple cameras is going to be absolutely epic, but that is a fairly hard problem.

[fspear]

I have a feeling that e-learning is going to be the killer app for VR/Augmented reality.