Wednesday, February 18, 2009

keyboard layouts, phonetic.

D: this group manually inputs pronunciation for dictionary words.
I am unsure why they don't just use a computerized version of the standard USA or UK speech guides, but meh...

"A Phonetic Alphabet for Keyboards

We needed a practical phonetic alphabet for use with a standard computer keyboard to annotate large numbers of words for our English pronouncing dictionary at howjsay. These phonetic annotations were needed as an aid during recording sessions.

Our need was for a phonetic transcription method that was fast, concise, and flexible so that it could be extended as needed where finer distinctions between sounds were required, but 'collapsed' where these finer distinctions were not required.

To increase typing speeds, upper case symbols were mostly avoided and the commonest items were assigned to easy-to-reach keys. New symbols are being added as the need arises, but so far (June 2007) this is what we have:"

D: Look at their suggested layout, heavily relying on punctuation keys and symbols:


Sound as in…

Vowel sounds
(defined as unstopped sounds)

sat, bat
sad, bad
hard, father
set, wealth
air, where
sit, faces
seat, receive
cot, watch
caught, ought
(numeral zero) coat, phone
cut, mother
put, foot
boot, food
German 'über', French 'tu'
about, problem
bird, her
French 'jeu', German 'schön'
her, Ohio
well (shorter than 'put')
yell (shorter than 'seat'
café [Eg kafe/]

D: The consonants are a much better match. English has sooo many vowel sounds compared to vowel letters. The Phoenician system no longer serves us well.

The ASCII Phonetic Alphabet

The International Phonetic Alphabet is very popular, but there is a big problem with this alphabet: the IPA symbols are difficult to type on computers. You can do it, but you need special fonts and special software. This is very inconvenient.

Therefore, when you want to write English sounds in computer documents, or in e-mail messages, or in SuperMemo collections, it is better to use a phonetic alphabet which doesn't use strange symbols like Z or @, but uses regular symbols like Z or @ instead.

We have created such an alphabet. We've named it the ASCII Phonetic Alphabet, because the letters and symbols displayed by computers are called ASCII characters. (By the way, "ASCII" is pronounced ['@s ki:].)

D: their system relies more heavily on the 'other' symbols on a keyboard. You know, the one most folks have no idea what to do with. Or what they are called. God knows I don't.

IPA ASCII examples listen
^ ^ cup, luck Amer
a: a: arm, father Amer / Brit
@ @ cat, black Amer
e e met, bed Amer
.. .. away, cinema Amer
e:(r) e:(r) turn, learn Amer / Brit
i i hit, sitting Amer
i: i: see, heat Amer
o o hot, rock Amer / Brit
o: o: call, four Amer / Brit
u u put, could Amer
u: u: blue, food Amer
ai ai five, eye Amer
au au now, out Amer
ou Ou go, home Amer
e..(r) e..(r) where, air Amer / Brit
ei ei say, eight Amer
i..(r) i..(r) near, here Amer / Brit
oi oi boy, join Amer
u..(r) u..(r) pure, tourist Amer / Brit

D: plus I found an IPA keyboard overlay. I considered retaining the layout for HIOXian.

The details about pitch and lip rounding I had planned to offload onto the diacritic.
D: apparently this one is uniquely well suited to typing Esperanto diacritics. The circumflex on consonants.

Aside: Antimoon's advice on learning English? Don't speak it - you might make a mistake.
I think they are right that unlearning an error is problematic.
But this terror of mistakes in nat-langs is one of the best pitches for an aux-lang.
Worse, a word that is literally correct may not be the right synonym to use.
I even read journalists that are unable to select just the right word for the turn of phrase they seek.

I pondered what early VERSE would look like as an English lesson.
1) teach the simple grammar rules
2) send home common irregulars and exceptions as homework assignments.
3) mention there are dozens, if not hundreds of exceptions in most categories.
D: their hopes are dashed. If only the rules of English were LAWS.
I feel sorry for ESL students. Just trying to parse word boundaries in English is difficult. Without parsing, there can be no vocabulary acquisition or learning rules of grammar.
Unpredictable stress, lax syllable construction rules, non-phonetic spelling. Colloquial deformation.
Once they can detect words (assuming our phonemes don't make them effectively half-deaf to English sounds) , then they encounter multiple homophones and the same word in different applications. Go and will are verbs. They are also auxiliaries in some forms.
Dead ends for rules. Invulnerable. Not able to... vulner?

Soon they wish English really did have laws and not rules. With some revisions, it can function with remarkably few laws.
And there you have it - VERSE - at first, a sort of con-lang. Later an aux-lang.

I think topicalizing is useful.
I noticed my writing often gets plural agreement wrong.
x and y (verb default to y as single).
Instead of
x and y, they (verb agrees with plural pronoun).
It is safe to say I have a hate-on for agreement.

No comments: