Related topics

Science Fiction's Predictive Ability?
It broke my heart to buy a Dell laptop for the single solitary purpose of using speech recognition. Sorry for the delay. I have been busy lately trying to talk my surgeons into installing a Lotsa Japanese names, repeated in profusion throughout his books, so using the 'spell' feature slowed me down too much.

Could Japanese adopt Hangul?
(One of these auto-completion feature invokes the spell checker to complete a word! That's cool.) cr88192> yes. the way most c code is structured would make it cr88192> nearly impossible for general speech recognition, though cr88192> it may be more possible with a lisp-like language, Yeah. a la Lojban. cr88192>

!! Intel goes APE-SHIT after hearing about the X86 Conso
With Vista speech recognition you can dictate to Microsoft Word, you can dictate to internet explorer, you can dictate to Outlook but not to Firefox or Thunderbird. Yet, the speech recognition program can communicate with these other programs. If you say a link in Firefox it will go to it, and you can spell out

Assistive Technology-Voice to text
The idea is to use a neural net (an unsupervised Kohonen feature map) to perform phoneme recognition, and then to translate the phonemes into graphical lip-shapes. This should be much easier then full speech-to-text recognition as there is no need to spot word boundries or spell correctly, the lip reader does the

Pathological Speech-Recognition Software
This is a common problem in speech recognition called a homophone error. In the current state of voice recognition software algorithms there is not yet This makes the utilization of spell check useless in detecting homophone errors. Generally speaking homophone errors can easily be recognized by a person in a

LShaping cannot spell or read
Accordingly, both spell checkers and grammar checkers will miss them. If I overlook even a few when proofreading, the reader will suspect was high on LSD while writing my report. I suspect the problem is with current speech recognition engine technology. We may be reaching the limits of statistical language models.

why won't speech recognition work fully with non-Miccrosoft pr
Would we even care to guess just HOW many words there are in the English language with which to train these recognition programs when we have to start with the simple examples you gave such as "or=all" & "an=and" ? You see, the trick to a really great "typist" is not that he/she knows how to spell Aspergillus

Character recognition
Some more general points:- Speech recognition is very hard work for the computer - you need lots of processor speed, lots of memory and a big hard disk. They never get the spelling wrong - just the whole word - so your spell checker won't pick up any mistooks. The dictation program "knows" a bit about language

Help With Speech Recognition Software
Speech recognition is not language dependent. The sound "phhttt...." can be made to translate into the character string "antidisestablishmentarianism" simply by training the program to associate There is no way such a program could spell "compter" with "computer" or "expert" with "exp[rt" in the same paragraph.

Question about SP2 and Spell Checker
I could read okay but I couldn't spell and my handwriting was incomprehensible. I spent two years in that intensive phonics program and didn't get much out They couldn't program a speech recognition system that could figure out individual word of the English language. If this poster's knowledge of things is on

practical book on NLP and speech recognition / sysnthersis
I am presently teaching my computer (Voice Recognition) to spell stuff Neandrathal style the way I talk and soon will not need the computer keyboard at all, but today was the first day of changing voice to type and see there is a long way to go. The program scans a letter that we read and then matches the speech to

Dyslexic Spell Checker
I read through a long thread on this NG a month or two ago in which people discussed the state of speech recognition and dictation software on the Mac and the any keyboard shortcut, run any applescript, do all cursor movement, do any cursor operation, dictate into any application, or spell into any application.

Put Googles adaptive spell check engine in Office.
I have not yet seen any voice recognition program that copes with a full range of natural speech even when trained and using very clear speakers. Finally, I saw a neurologist on Two specific problems are:- It never gets the spelling wrong - just the whole word - so your spell checker won't pick up any mistooks.

IBM's MedSpeak/Radiology Continuous Speech Software...
Obviously speech recognition is a very important technology, but there is little chance that keyboards will disappear anytime soon (for a start, you'd have to spell out every word the system didn't know letter by letter and that's often hard enough to do with a person let alone a machine).

Say "wreck a nice beach" quickly
The downside of speech recognition programs is that they never misspell a word! That sounds like an advantage but it makes it impossible to find errors using a spell checker because the kind of errors it makes are things like "what" instead of "put". Some of the errors come out really funny!

MWNY shows Apple targetting high-end consumers
Speech recognition is still far from practical, although it =would be interesting to see how well the AV Macs do; Sorry, there are speech-recognition systems The robot in that story was basically a spell-checker and grammar-checker. The premise of the story was that someone opposed to such things tried to make

speech recognition
It should be recent enough to cover modern research areas as well. at the moment im reading the book " Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics and Speech Recognition " by Daniel Jurafsky and James H. Martin, but im not sure whether the book is mostly

Newton Dragon Speech
Try the following:- - buy a very expensive PC, - add speech recognition software, - spend a week or so training it to recognise "The Master's Voice", - print out the results, - scan into Acorn and use OCR software, - spell check. Hey presto, you get "Send three and fourpence, we're going to a dance!

Xenu Agent: Solution looking for a problem!
No price listed Lernout & Hauspie Voice Express Pro www.lhs.com Continuous speech recognition MS office and other Windows apps compatible Control and command difficult to make the sucker change it to the "right" word - you have to spell it using military alphabet until the program can finally guess the word).

Accurate Speech Recognition? Ever? Never?
Voice recognition attempts to identify a speaker by voiceprinting and speech recognition attempts to identify what words are being spoken by the speaker (otherwise known as "speech to text" translation). "This peach wreck ignition soft where will knot spell czech write ether"