|
-
Oct 19th, 2011, 02:36 AM
#1
Thread Starter
Banned
Re: The iPhone 4S
some questions regarding siri's voice recognition :
1 is the algorithm based on words & sentences sound to text or phonemes to text ?
2 the sound data processed, is it the time between 0 volume levels(volume sample time) and volume peaks per volume sample time ?
3 what is the interval in which a sound (volume, buffer) taken ?
4 does the voice reco have a mistake reduction mechanism ? if yes, plz if possible, tell me some about it's algorithm
5 and most importantly, from where have the codes (or techniques) that are used were aquired ? and could they be used in vb.net ?
granted win 7 has voice recognition but it is difficult to train as the spelling recognition correction uses
the user voice, unabling the access to too many letters, at times
Last edited by moti barski; Oct 19th, 2011 at 02:40 AM.
-
Oct 19th, 2011, 03:05 AM
#2
Frenzied Member
Re: The iPhone 4S
 Originally Posted by moti barski
some questions regarding siri's voice recognition :
1 is the algorithm based on words & sentences sound to text or phonemes to text ?
2 the sound data processed, is it the time between 0 volume levels(volume sample time) and volume peaks per volume sample time ?
3 what is the interval in which a sound (volume, buffer) taken ?
4 does the voice reco have a mistake reduction mechanism ? if yes, plz if possible, tell me some about it's algorithm
5 and most importantly, from where have the codes (or techniques) that are used were aquired ? and could they be used in vb.net ?
granted win 7 has voice recognition but it is difficult to train as the spelling recognition correction uses
the user voice, unabling the access to too many letters, at times
I'm not sure anyone is going to be able to answer those questions. All I know is it understands context so it must be processing the words and figuring out what is meant. Additionally, from what I've read (unverified so far) it uses one of the cores in the SGX543MP2 for background noise filtering and to home in on vocal structures, as well as breaking the string down and sending it back to Apple. Most of the work is done on Apple's server.
No one knows any of that stuff yet... I'd keep an eye out for some thorough reviews of Siri, perhaps from Anandtech... they generally seem to do good and thorough reviews and I've no doubt Siri will get plenty of attention. Not sure details such as what you've asked for will be covered though.
Tags for this Thread
Posting Permissions
- You may not post new threads
- You may not post replies
- You may not post attachments
- You may not edit your posts
-
Forum Rules
|
Click Here to Expand Forum to Full Width
|