Text -> Speech: If you see a paragraph containing unknown words, you should be able to read it aloud, without any mispronunciations.
Speech -> Text: If you hear a paragraph of spoken text containing unknown words, you should be able to write down the words, without any spelling errors.
Once, when my Brazilian wife was quibbling about all of our funky spelling in English, she said how everything is pronounced exactly as spelled in Portuguese.
Unfortunately, for Americans learning Portuguese, it ain’t so:
In the speech->text direction, there is ambiguity with “ss” and “ç” (among others).
In the text->speech direction, there is ambiguity with open vs. closed vowels: “colher” could be “spoon” or “to pick”, depending on the vowel sound.
Spanish seems to be better, but I have never heard any Spanish speaker pronounce “b” and “v” differently. (“v” de “vaca”?)
Spanish comes really, really close in the text -> speech direction. I can pick up something written in Spanish and say it out loud with no problems, whether or not I’ve seen the words before. There are some Spanglish words that they’ve borrowed (like el look or el show) that don’t match the orthography, but that’s about it.
However, there are some difficulties in the speech -> text direction.
This is one of them. B and V represent the same two sounds with no distinction between which letter goes with which sound. The soft C (like in parecer) and Z sound the same, and in American Spanish, they sound the same as S. The H is never sounded and is left over as a relic from days gone by. As a result, a ver and haber sound exactly the same. I’m sure there’s something else I’m missing too.
Even with these things, it’s certainly more logical than English.
I might be misremembering it, but I recall German being quite rigid regarding the rules of pronunciation, for example, ‘ie’ is always pronounced ‘e’ and ‘ei’ is always pronounced ‘i’, and the language isn’t frightened of adding syllables, so new words formed by conjoining two existing ones often just stick the two words together without altering them.
But it’s been some years since I used the language, so I might be forgetting something.
Esperanto, perhaps, if we could come to agreement on a standard pronunciation at all. But the notion of one letter one sound is there…
The fact that it has never caught on is probably another lesson here about any intrinsic value (or possible success) standardidization has in language.
If we can include planned languages, Esperanto is pretty damn unambiguous (probably because it was planned to be that way). Every letter has exactly one pronunciation (except for certain diphthongs, but they could simply be considered as letters unto themselves), letters don’t share pronunciation, and stress always falls on the second-to-last syllable.
There could be a bit of confusion in the speech->text direction though; ‘c’ is pronounced pretty similarly to ‘ts’, for example.
A third for German. From my high school german classes I remember the rule that once a word is written down there is only one way to pronounce it correctly (excepting dialects etc.), and also that there are no “silent” letters, many that if you remove any of the letters the word would be pronounced differently. Which would fit in with the German stereotype of exactitude - or maybe their national character was fashioned by the language!
Italian? Upon learning the basic rules of pronunciation and spelling, I’ve never had any problems. Text->speech definitely works & speech->text works if you aren’t horrible at spelling (the written “h”, signifying a hard consonant, being a good example of this). You pronounce every letter you’re given, stress is non-ambiguous, etc.
Loglan, at least in its early form. Not only is the phoneme set small, related strictly to the alphabetic representation and chosen for widest possible variation without ambiguity, but it is designed for an exact match between the written and spoken forms, with nothing conveyed by typography and punctuation that isn’t spoken, and nothing to be conveyed by spoken inflection that isn’t written. For instance, the concept of quoting is indicated by operator words which are identical in the spoken form, not by quotation marks. You could theoretically speak all Loglan in a flat monotone if you wanted to, without losing any meaning.
There are five vowel sounds
a as in “car”
e as in “meh”
i as in “ski”
o as in “door”
u as in “tutu”
two different vowels together form a dipthong with sounds according to the above scheme.
Two vowels the same together is a repeated sound – two syllables with a slight pause between.
All syllables are either stand-alone vowels or a consonant sound followed by a vowel. No syllables end in a consonant sound.
There are 10 consonant sounds:
H, K, M, N, NG (as in “sing”) P R (wich is slightly rolled) T W Wh (pronounced “f” in most old dialects and in modern maori)
Thus there is an easy phonetic connection between what is written and what is spoken. If you pronounce it correctly then you can spell it. If you see it written then you can pronounce it.
The only ambiguity is which syllables have emphasis placed on them. In general there is an accent on the first syllable of a word and a secondary accent on the third. But this does vary a bit.
I’d also say Italian. Every word is spelled exactly how it sounds.
Of course you have to know the difference in sound of a single and double consonant, and the other rules of how certain sounds are spelled, but if you do, then every word is easy to spell and you should be able to write any Italian you hear correctly, even if you can’t speak it.
100 percent phonetic. Accent is ALWAYS on the first syllable. No genders and no articles. Of course the prepositions are all suffixes, resulting in some thirteen oir fourteen cases.
Examples:
Talo=house
taloon= into the house
talosta=from the house
talossa=inside the house
talolla=at but not inside the house
talona=as a house (the building was being used as a house)
taloksi=becoming a house
And so on. learn the prepositions and you have the language knocked.
Oh yeah, forgot about Maori. Once you know the simple rules, impressive words like Taumatawhakatangihangakoauauotamateaturipukakapikimaungahoronukupokaiwhenuakitanatahu are a cakewalk.
One problem is that we need to define what “mispronunciation” means. There are a large number of phonetic or near-phonetic alphabets; syllabaries tend to be very good at this, and the original version of Hangul actually described sounds in terms of their place and manner of articulation.
However, these orthographies (writing systems) tend to miss the allophones, which are multiple sounds which don’t create lexical contrast in a given language. Aspiration in practically every dialect of English, for instance, is an example of allophony: put your hand in front of your mouth and say "truck "a couple of times, then say “tip” a couple of times. You’ll notice that you can feel air blowing over your hand after the “t” in “truck”, but never when you say the “t” in “tip”. This is because the truck’s “t” is “aspirated”, that is the release of the “t” sound is followed by a strong burst of air. In English, if you pronounce “truck” without aspiration or “tip” with aspiration, it’ll sound a little funny but you’ll never change the meaning. In other languages, however, aspiration is contrastive: in Tibetan, for instance, aspirated and unaspirated consonants are considered to be different sounds.
Oh yes, that’s why I quoted the Japanese bit; if a native speaker sees hiragana or katakana they’ll pronounce it like the other speakers who grew up in their speech community. However, they’ll include some variations that the plain orthography doesn’t represent. Whenever two stops occur adjacent, for instance, they tend to realize the inital stop as a velar nasal.
To add a bit on Japanese syllabaries (kana), the current usage is the result of a fairly recent standardisation, but some archaic irregularities remain:
は : can be pronounced /ha/ or /wa
へ : can be pronounced /he/ or /e/
A native reader would never mistake one pronunciation for another, but getting it right requires knowledge of the language. Furthermore, the kana う indicates a long vowel when it follows an /o/ sound. For instance, Tōkyō is とうきょう. However, in some cases, long /o/ sounds are denoted not by う but by お as in Ōsaka: おおさか. There’s no easy rule to figure out which are the few words that are to be written with お instead of う.
There’s another issue I see and that is dialect: a phonetic representation might be accurate in one area but not in another. Using long-short vowels in Japanese as an example again, grandfather is おじいさん(o-jiisan), the /i/ is long, which is denoted by the kana い. This is important to distinguish it from the similar おじさん (o-jisan), which means uncle. The word for “tree” is written き in hiragana, which is pronounced /ki/, a short vowel. However, this is the eastern pronunciation, around Osaka, the word is pronounced with a long sound. Some writers sometimes write it as きい or きぃ to stress the Osaka pronunciation but it is not standard orthography.
