Voice Recognition

Hey guys, I don't think it's ever really been clarified about how exactly the Voice Recognition will work on the Mk2's. Is this purposeful or is it still under development?

My dream would be being able to say "empeg. Less Than Jake. Shuffle." and it play all my LTJ tracks shuffled, or "empeg. Radio." to switch to radio mode. Maybe even "empeg. Tune to 93.3.".

How far-fetched is this? Or how is it going to work?

/me is getting antsy as June approaches.....

(O|||||O)

Has this question been answered already? Why hasn't anyone replied to him? Did I miss the explanation somewhere?

The developer primarily responsible for Voice Recognition on the car player is currently locked away working on another project with extremely tight deadlines. He probably doesn't have much time to read the BBS at the moment!

We'll announce more voice recognition details nearer the launch of the Mark 2.

Rob

The developer primarily responsible for Voice Recognition on the car player is currently locked away working on another project with extremely tight deadlines.

Make sure to slide some food and water under the door for him once in a while.

Tony Fabris
Empeg #144

Make sure to slide some food and water under the door for him once in a while.

---

...only drop some caffiene pills in his water...
-m

...proud to have one of the first Mark I units

Pizza is of course the ideal food for this since it comes in a flat box that can easily be inserted under the door. I find that beer or a caffeine based beverage are actually more usefull than water when it comes to enhancing programming skills. Unfortunately a pint of Guinness can be a little harder to fit under the door :-)

-Mike

Less than Jake? That sounds utterly weird!

How about...

Empeg! Classical Music, baroque style, random artists, lively pieces only.

Empeg! Pause! Now play the next 10 seconds in reverse repeatedly.

Empeg! Play this current song only through the left front speaker. Play the next song in the playlist through the right front speaker. Mix them to sound good.

Empeg! Adjust for road noise.

Empeg! Play all the songs that sound like this one.

Empeg! Remove the drum track from this song.

Empeg! Karoke mode.

Empeg! I'm going to take a nap, take over the steering.

Calvin

Wait 20 years, and this will probably work.

About the only reason I haven't committed suicide yet. ;)

Empeg! Order me a pizza, delivered to the car.

Alex

I really think it would be corny and monotonus to keep saying EMPEG! ..
I'd much rather say just Play Some Hip-Hop and it would play my hip-hop playlist.. or Play The Hot Shit.. and it would play my playlist "hot Shit!"

I hope you make it configurable where u can change the voice command to suit you.. So i can make it as complex or as simple as I want it.



Empeg Kicks Ass
S/N 00203
http://www.iretro.com

One of the problems with a voice recognition computer is that you don't want it to respond to every piece of conversation you have. An in-car computer can't tell whether you're talking to it or talking to the person in the back seat. Hence the need for a command prefix. I don't know if the Mark 2 will need one or not, I'm just saying that's why everyone assumes we'll have to say "Empeg" before each command.

I always thought this was an inconsistency in Star Trek. When talking to the computer, they would prefix their commands with "Computer". But when they got onto the elevator, they would just say "Deck ten". How does the elevator know that they're not just having a conversation about the sexy young ensign they just met on deck ten?

Tony Fabris
Empeg #144

Tone and inflection. Current voice recognition systems (of this type) require a command word to clue in that it should be accepting commands. Future voice recognition systems will rely on tone and inflection. The same way when you talk to a baby in a "baby voice" it understands you are speaking to it. Or when you're talking to a dog, other humans around you, even blind humans, will not mistake it for a conversation! Some day computers will do the same, and humans will adopt a "command mode" voice for interfacing with computers. I read some early prototypes are using "whispering" voices to indicate command mode.

Calvin

i would prefer the option , to just have it interpret it from regular conversation whenever i want it to. rather than be stuck saying EMPEG. I want the option to switch the command preface to EMPEG , or Sharon , or anything i wish.. or nothing at all.

I think that a choice is extremely important. That's why im trying to mention it now before they make a hard-coded version which i think is undesirable. As a consumer , i want as many options as i can get. In software implementations , i know that many options are configurable if they develop it that way.


Empeg Kicks Ass
S/N 00203
http://www.iretro.com

Keep in mind that they're licensing the voice recognition from a third party. There's a possibility that Empeg won't have much control over how you have to talk to it.

Tony Fabris
Empeg #144

The speech recognition API in the MacOS does this in what I think is the most logical way -- you give the software a command dictionary of words or phrases that it should understand, and it just lets you know when it hears one of the commands it knows about. Works pretty well, too.

IMHO, this is the way the Empeg SHOULD work, since this would prevent having to record each of your playlist names for it to compare -- it should know your playlists, and be able to do speaker-independent recognition based on that. Again, since you're not doing continuous dictation of arbitrary works, reliability is good.

Alex

The Macintosh isn't designed to work in a high noise environment either. You'd be surprised how badly a generic voice recognition can be with a high enough noise level, for example, a car rolling along a highway.

Calvin