Re: [AfricanLanguages] Re: Recordings with Transciption

[paa kwesi imbeah]

Purely speculating in the following muse:

I think the advantage of writing is as a memory aid--assuming a finite memory recall ability, it is more efficient to recall an index of texts than the contents of the text itself. Writing allows us to file away detail and only have to remember that "there's a book on the subject" whenever we need to recall that detail.

Orality doesn't allow for that since there's no stored memory apart from the collective memory. For there to be a "neo-oral civilization" with comparable recall ability to literate civilizations, we should be able to store massive amounts of oral knowledge organized in a sensible index ( e.g. the title of a book is an index to the contents of the book). And this index must itself be oral in order to be accessible as accessible as its content. The closest I've come to seeing this idea in practice is the interactive voice response system where you can call a telephone number and find out answers to particular queries by following a series of oral prompts. 

It would be nice if the equivalent of paper for audio were invented so that storing audio is a simple as writing on a scrap of paper.

End of muse.

More realistically, the reason I'm interested in spell-checking is to make a workflow like the following possible.

English text => Standardized African language => Variant languages => speech from text. Computers unfortunately are incredibly dumb and can be thrown off more easily by spelling inconsistencies than humans.
50 cedis,
paa.kwesi

On 9/19/06, Don Osborn <[EMAIL PROTECTED]> wrote:

Hi Paa Kwesi, I wonder if ultimately we might not turn the technology
more fully oral so that it can check the sounds, while accounting for
dialect differences and match the WYSIWYG orthographies to that. Text
has, since the beginning more or less been something fixed in stone,
clay wood or ink. This has had a lot of advantages for learning and
unambiguous transmission of messages. But is there a degree to which
the fluidity made possible by computers and bytes can make the
regularity of spelling less important for those objectives than the
trueness to speech and meaning?

What if a computer could "read" (TTS) a text and match the sounds with
a sound corpus with meanings (it is possible to search sound strings
in digital audio files, so I understand, so this would I think push
the matter ahead a couple of degrees of sophistication). Basically do
a spell and grammar check as audio internally rather than as text
strings. A computer could "read" this way much faster than we can - as
computers are made to do everything more rapidly.

It's been a long day and this is off the top of my head, but some of
us like Tunde Adegbola have chatted about how ICT can make possible
literate + neo-oral civilization in Africa. Not sure what this might
mean, or if it's even practical, but the possibilities opened by ICT
are not, of course, limited to repeating what others have done with it.

All the best.

Don

--- In AfricanLanguages@yahoogroups.com, "paa kwesi imbeah" <[EMAIL PROTECTED]>
wrote:
>
> Forward march towards ubiquitous spellchecking ability...
>
...

--
www.kasahorow.com > communicate __._,_.___

SPONSORED LINKS

Creative writing	Creative writing course	Creative writing class
Creative writing program	Teaching creative writing

Your email settings: Individual Email|Traditional
Change settings via the Web (Yahoo! ID required)
Change settings via email: Switch delivery to Daily Digest | Switch to Fully Featured
Visit Your Group | Yahoo! Groups Terms of Use | Unsubscribe

__,_._,___