Thread: DFX 5.1.2 Released [ Updated 04-02-12 ]

Blue, This would be cool to add. Please add this cmd to the mix. I'll never have to tap on screen to Maximize/Restore from being Minimized when I'm using RRVoice. Thanks in advanced bud.

added "restore"

course, the only way to activate it, would be from outside influence
rrvoice is good example

more i think about it, RR shouldnt have minimize!!
i see no real reason to leave the FE
its just not right

Well, first try to actually own/use a carpc, then we'll go from there...LOL

Originally Posted by Blue ZX3

Well, first try to actually own/use a carpc, then we'll go from there...LOL

Ouch!

Originally Posted by Blue ZX3

Well, first try to actually own/use a carpc, then we'll go from there...LOL

Boy, did that hurt to do?

Originally Posted by Blue ZX3

Well, first try to actually own/use a carpc, then we'll go from there...LOL

What!, Mitch you don't own a carpc??!! Say it's not true.

Mitch thanks for the adding this option buddy.

has anyone else had this idea for rrvoice?

a lot of times when im asking to play a certain album or song, i cant always remember the exact name. since i dont say it right, rrvoice just sits there in listen mode waiting for me to say it again.

what if rrvoice instead searched for the closest two songs that matched words i said, and asked me what one i wanted? so you say the album wrong, but it asks you "did you mean this album, or that album?" then it would automatically wait for my response, and since it read some examples i would be able to say it the right way.

i would think this could be implemented for a lot of things... just having it ask me questions instead of sitting there like an idiot would be really nice. i know my truck is not an idiot but sometimes man... sometimes....

no i dont own a carpc

Originally Posted by trader007

has anyone else had this idea for rrvoice?
What if rrvoice instead searched for the closest two songs that matched words i said, and asked me what one i wanted?

The way Voice the voice recognition in the plugin works, this can not be done. A dictionary is built from the pre-defined phrases.

It would take open dictation to do that, and that would be HIGHLY inaccurate.

For instance, the song "Paradise by the Dashboard Lights" is a phrase in the dictionary, if you used dictation, it could think it heard something like "raster size by the flash guard mights"

Using a phrase, when you say that song title, it basically compares it to all other search strings, and picks the one it is pretty sure you meant to say. So when the sentence is matched as a whole, it is far more accurate.

I have found over the years, that dictation works about 80% on the number of syllables in a phrase to match by. So if you have "Turn on the lights" as a command, it would be best to not have a command "Turn off the lights", but instead, use "Turn the lights off". This gives a distinct difference for the voice recognition comparison to choose from what it thought you said.

As an experiment, when the system is listening for a command, say the command much much faster than you normally would. If you pronounce all of the syllables correctly, it will most likely still understand you. Speaking faster or slower really seems to have no effect on speech recognition in windows.