mastayoda

Great job! Thanks for sharing this!

I'm developing a dynamic tutorial desktop app and I plan to use this model as text to speech synthesizer. Any chance it can be ported to ONNX format?

Something similar happened to me! Pretty much after booking and waiting for the checking instruction, the guy asked me to go 1 mile away to pick up some keys in a unknown location, that was not specified in advanced... I wanted to cancel the reservation to pick up a more safe place... Airbnb only return 50%... this was in matter of minutes! The guy wanted 50% for a reservation done 30 minutes before sending me those ridiculous instructions... I contacted Airbnb, scaled the situation, and at the end they said nothing could be done... yep, they are enabling Scammers... Never used Airbnb again...

Damn right!! on the path to make actually do what it is supposed to do.

How many years took them to realize this?... As someone who spend like 6 months job hunting and doing more than 20 interviews(including Google, Microsoft, etc), this was pretty evident. Yeah, I forgot to code that weird problem which implies the use of dynamic programming in a white board, but hey... what about all those years of experience and evident code skills. It is ridiculous.

I have good and bad(maybe not bad at all) news.

- The good one is that this is a great and already validated idea. There is a plenty space of opportunity in this and it is indeed a missing piece in the AR space. So I would tell you go ahead!

- The bad news(not so bad) is that you have strong competitors: The company I work for(Hologen.io) have this service implemented already. We cover ARKit, ARCore, and Windows Mixed Reality. It still not in production yet but we expect to release in the upcoming months. As I tell my team all the time, the more AR products and ecosystems, the better, so even if we are competitors the proliferation of these systems helps all of us as well, so go for it!