Pogoplug project idea.

I picked up several pogo-plug mobiles because they are really cheap and I had some success in the past using one as a NAS. But now I'm wondering, with all the people using them as webservers, etc. Why not try something more creative? Here's my idea - I'm wondering if the USB port could be repurposed as a sound output and a sound input device? Particularly, utilizing API calls to harness cloud processing power and using the POGOplugs to run input and output (E.G. one dedicated to microphone and one for speaker).

There's the Google Speech API - Converts audio to text in 80 languages.

ALICE - Opensource artificial intelligence "chatbot" which can be fed text via API and output responses.

Then Google has the Cloud Speech API which does the opposite of the input - and it actually sounds pretty natural.


So basically, I'd like to mount it to my front door and use it to intercept people selling magazines and things like that when I don't feel like talking to them (see illustration).

I'm pretty new to linux/debian so I'm trying to figure out what software would be best to accomplish what I'm trying to do.

Quote

"When I was a little boy, I used to pray every night for a new bicycle. Then I realised, the Lord, in his wisdom, doesn't work that way. So I just stole one and asked Him to forgive me." - Emo Phillips

GitHub Profile



Edited 1 time(s). Last edit at 12/02/2016 02:33AM by november17.

yes its possible - have a look at bodhi's wiki page for starter information

That's a lot like what I've wanted to do with an AI answering machine. Phone rings, you see the source, and if unknown, hit the Dr Phil button. Then Dr Phil talks to them. So it would take a sample bank of the sayings of Dr Phil (my pet peeve deluxe), and it would involve a flowchart of questions from the good Dr.

The idea for me goes back to when there was (maybe still up) a website with a page full of sayings from famous people. Not too many words cover a lot of ground.

But there's something much more valuable for you and me: many companies do anything they can to get you to say "yes" and when they have that recording, they can do bad things you know. they can say you agreed to certain services and whatnot. So if the only voice is Dr Phil or Richard Simmons or Donald Trump, this could be great fun.

Having said that, Adobe Project VoCo is Photoshop for speech. It will analyze a person's voice and you can make up new words out of it. So this could feed the speech library with real interesting items.

=========
-= Cloud 9 =-



Edited 1 time(s). Last edit at 12/25/2016 12:17PM by JoeyPogoPlugE02.

> So if the only voice is Dr Phil or
> Richard Simmons or Donald Trump, this could be
> great fun.

Yes, that voice would be: YUGE! :))))

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

bodhi Wrote:
-------------------------------------------------------
> > So if the only voice is Dr Phil or
> > Richard Simmons or Donald Trump, this could be
> > great fun.
>
> Yes, that voice would be:
> [url=https://www.amazon.com/gp/product/1449481337/
> ampweb-20]YUGE![/url] :))))

LOL and a half :-) If i get enough vocabulary I can IP-call Rosy O'Donnell // naw naw, probably bad idea, but funny as heck

=========
-= Cloud 9 =-

november17 Wrote:
-------------------------------------------------------
> There's the Google Speech API - Converts audio to text in 80 languages.
>
I sure would like to see if any Pogoplug has enough power to process this audio-2-text.