What do you think of phonetic transcriptions?
Every too often I find what people funnily call "phonetic transcriptions", where words like "presente histórico" are written like "Preay-sayn-tay ees-toh-ree-coh". A native speaker of English with no knowledge of Spanish pronouncing that gibberish sounds funny at best. Here you can see the spectrograms of these two phrases pronounced by a native accent above, and with American accent using the "handy" phonetic transcription below (yellow line is intensity; blue line is pitch):
(A higher definition one can be found here)
Surely you've noticed those dark thick ripples in the native's spectrogram. Those represent the sustained tension with which Spanish vowels are pronounced, as opposed to the wavy glides of the ever changing English vowel sounds. The differences in pronunciation are remarkable. Only the S seem to be similar. You should listen to the audio side by side, hahaha! It is hard to believe they are saying the same thing.
Now, why is the IPA (International Phonetic Alphabet) so unpopular? Is it because it is hard to learn? For those who don't know, it allows to represent phonetically (phonetically for real) the sounds, intonation and stress of most languages with a reasonable degree of accuracy. Each symbol has one sound. It is 100% regular, making Spanish look like random in comparison. Although some symbols look funny, some of them are rather easy, at least in Spanish. This is how to represent "presente histórico" with the IPA:
/pre'sente is'toriko/
The vowels in the IPA look like the vowels in Spanish (even though this system is British), but for English you need at least 15 symbols to accurately represent the sounds. The phrase "Dalai Lama" has four "a" written in it, but it contains 4 different "a" sounds: /æ/, /a/, /?/ and /?/.
The IPA allows you to see subtle differences, like how the two B in "bebo" sound different: /'be?o/, where the sound /?/ is pronounced without closing the lips (it doesn't exist in English).
Even I can pronounce correctly the name of the volcano "Eyjafjallajökull" in Iceland if I am given the IPA pronunciation (/???ja?fjatl?a?jœk??tl?/). I have surprised people in many exotic languages reading their words when I had these proper transcriptions in front of me (although I read a bit too slow, unfortunately).
This system is extremely useful for English, for example, where you can check the pronunciation of an unusual word, like a weird chemical compound or enzyme that no one can pronounce, but with phonetic languages like Japanese, Italian, Turkish and Spanish has only a limited practical use if you are not familiar with the sounds. It is, however, a very systematic way to approach sounds for thousands of languages, so I think it is worth learning, if you are curious about how languages sound.
A lot of articles in Wikipedia (thankfully) provide the IPA pronunciation, which is handy if you don't know how to read the word in English, or when it is a foreign word. A few symbols, and you can tell how the English and the native pronunciation differ. It is an essential tool if you love to know about foreign sounds (especially when no sound clips are available).
This IPA was created at the end of the 19th century, and although it was not been used much in the USA, it is fairly common in Europe and other places like China.
Does anyone know this system? Has anyone heard of it? Do you think it is pointless? Would you like to know more? I'm curious to see what you think about this.
15 Answers
I first learned about the IPA in a linguistics course in college (university) and thought it was a great idea. I briefly applied it to my study of Spanish but found it of little use (basically, for Spanish, it's overkill) mostly because I had always found Spanish pronunciation to be easy and the spelling to be so consistent.
A few years later, however, when I was studying Arabic, it came in very handy. In that course the introduction of the script was intentionally delayed for about a month and English letters were totally inadequate for making notes in class about pronunciation. I also found using the IPA to be helpful when studying Chinese.
With Japanese, like Spanish, the pronunciation is so simple and regular that using romaji or kana is the easiest solution.
During one trip to Japan, I did a short "home-stay" with a high-school English teacher and sat in on several classes. I saw the students all industriously making their pronunciation notes using katakana (a syllabic writing system that works quite well for capturing the sounds of Japanese but is completely inadequate for English sounds). In effect the students were keeping notes on how the English might be pronounced if one had really bad pronunciation and then going home and studying these distortions and, thereby, reinforcing bad pronunciation. This, of course, is quite similar to trying to capture Spanish pronunciation with English "equivalents". One ends up practicing how to mispronounce Spanish. This approach forces students to focus their attention on the sounds of English which are so often different from those of Spanish.
The IPA may appear to pose a steep learning curve but one needs only a tiny subset for studying any particular language (about half a dozen for Spanish sounds that are not represented by letters in the Latin alphabet).
I've heard of it, of course, and seen it used, but have never learned it. To me it looks a bit like learning Kanji or Arabic. I doubt it will ever gain much of a foothold except in the academic environment.
I have books over 40 years old, published in England, Spain and other European countries with this system, which was created in the 19th century. When my wife began to learn English in China, she was taught this system in her first lesson, and her pronunciation is way better than mine.
Lazarus in view of that observation how then can such a system possibly help people learning languages when it appears so inaccurate? I would be interested in knowing the answer to this mystery/puzzle.
The second recorded is based on the "inacurate" pseudo-transcription "preay-sayn-tay ees-toh-ree-coh" which I tell everybody to avoid. If the IPA transcription had been used instead (/pre'sente is'toriko/), the main features of the spectrograms would look very close. As a comparison, the same sentence pronounced in Italian, which is close to Spanish, but not as close as the IPA transcription, already looks almost identical to the one pronounced by the native.
There are two ways to learn a language: listening to a native speaker (the ideal way) or using a proper phonetic transcription, and "preay-sayn-tay ees-toh-ree-coh" should only be used as a desperate attempt to communicate if you don't speak the language, but never if you intend to learn it. The IPA has another advantage: some people, even though they listen to natives, they fail to differentiate sounds, tones and other features which do not exist or are not important in their language (i.e. the distinction between B and V for a Spanish native when they speak English, or the difference between the soft B and the hard B for an English native trying to learn Spanish). A little bit of knowledge of phonetics can help you, at least, realize that such differences do exist. If you got a good hearing and you imitate them good, then this IPA is probably not for you.
I personally wish that all those travel phrasebooks had the IPA instead of pseudo transcriptions, so I can say a few survival sentences if I go to another country where they speak a language I know nothing about. Instead of sounding like a drunkard gargling, I'd be able to read the sentences slowly, but crystal clear.
I became familiar with this system in college when I lived with a girl who studied linguistics. I have to say I found it difficult to read then, but that's probably because it was very unfamiliar and I didn't take the time to learn the sounds. She used to record me speaking and would translate what I said into IPA. I was amazed how many different vowel sounds I made while I spoke!
I would love to know more about it, especially if it can help me improve my Spanish pronunciation (which is atrocious).
I've found this post very interesting, Lazarus.
I think it's very useful, but also very difficult:
..it allows to represent phonetically (phonetically for real) the sounds, intonation and stress of most languages with a reasonable degree of accuracy.
... that's the reason why it's so hard to learn.
It seems to me highly recommended for those who want to learn more than one language. The more interest you have in learning languages, the more benefit you'll get from tools like this.
Anyway, I still think there is nothing like listening to and imitating the natives.
A last note to say that, probably, these 'pictograms' were recorded by "voice models" in order to highlight how different they look. I suppose that reality is a bit more caothic.
I am interested in it. But every time I have encountered the IPA at wiki, it has sounded very confusing to me. Although, the articles were generally about high-level linguistics rather than the IPA itself. So, maybe I wrongly associated the IPA with difficulty and confusion based on the content of the articles in which I found the system.
I learned the IPA as an undergrad and used it then, during graduate school, and throughout my professional career as a speech-language pathologist. How would we use it here on SpanishDict?
It does accurately represent vowel and consonant sounds. It also marks the stressed syllable. 
so I think it is worth learning, if you are curious about how languages sound.
I am very interested in learning this.
I wonder how useful it would be for Spanish where most words sound just like they are spelled.
I have never learnt the IPA alphabet - (only academics know it - native speakers do not learn it in order to speak English) but it is needed to overcome the illogical nature of English pronunciation / spelling.
In my opinion it is almost redundant for Spanish and in most other languages.
One can find sites on the internet to type these symbols.
IPA Vowels
? cup, luck Amer
?: arm, father Amer / Brit
æ cat, black Amer
e met, bed Amer
? away, cinema Amer (this one actually has a name, it is called a Schwa
?:? turn, learn Amer / Brit
? hit, sitting Amer
i: see, heat Amer
? hot, rock Amer / Brit
?: call, four Amer / Brit
? put, could Amer
u: blue, food Amer
a? five, eye Amer
a? now, out Amer
e? say, eight Amer
o? go, home Amer
?? boy, join Amer
e?? where, air Amer / Brit
??? near, here Amer / Brit
??? pure, tourist Amer / Brit
IPA consonants
b bad, lab Amer
d did, lady Amer
f find, if Amer
g give, flag Amer
h how, hello Amer
j yes, yellow Amer
k cat, back Amer
l leg, little Amer
m man, lemon Amer
n no, ten Amer
? sing, finger Amer
p pet, map Amer
r red, try Amer
s sun, miss Amer
? she, crash Amer
t tea, getting Amer
t? check, church Amer
? think, both Amer
ð this, mother Amer
v voice, five Amer
w wet, window Amer
z zoo, lazy Amer
? pleasure, vision Amer
d? just, large Amer
.......
1 Almost all dictionaries use the e symbol for the vowel in bed. The problem with this convention is that e in the IPA does not stand for the vowel in bed; it stands for a different vowel that is heard, for example, in the German word Seele, or at the beginning of the e? sound in English. The “proper” symbol for the bed vowel is ? (do not confuse with ?:. The same goes for e? vs. ??.
2 In ?? and ?:?, the ? is not pronounced in BrE, unless the sound comes before a vowel (as in answering, answer it). In AmE, the ? is always pronounced, and the sounds are sometimes written as ? and ?.
3 In AmE, ?: and ? are one vowel, so calm and cot have the same vowel. In American transcriptions, hot is written as h?:t.
4 About 40% of Americans pronounce ?: the same way as ?:, so that caught and cot have the same vowel. See cot-caught merger.
5 In American transcriptions, ?: is often written as ?: (e.g. law = l?:, unless it is followed by r, in which case it remains an ?:.
6 In British transcriptions, o? is usually represented as ??. For some BrE speakers, o? is more appropriate (they use a rounded vowel) — for others, the proper symbol is ??. For American speakers, o? is usually more accurate.
7 In e?? ??? ???, the r is not pronounced in BrE, unless the sound comes before a vowel (as in dearest, dear Ann). In AmE, the r is always pronounced, and the sounds are often written as er ?r ?r.
8 In American English, t is often pronounced as a “flap t”, which sounds like d or (more accurately) like the quick, hard r heard e.g. in the Spanish word pero. For example: letter. Some dictionaries use a special symbol for the flap t. special symbols
IPA what it means ? The vertical line (?) is used to show word stress. It is placed before the stressed syllable in a word.
For example, /?k?ntrækt/ (noun - contract) is pronounced like this, and /k?n?trækt/ (verb to contract) like this.
Word stress is explained in our article about phonetic transcription.
? ? is not a sound — it is a short way of saying that an r is pronounced only in American English.
I've heard of it, of course, and seen it used, but have never learned it. To me it looks a bit like learning Kanji or Arabic. I doubt it will ever gain much of a foothold except in the academic environment.
Very intriguing, Lazarus! I am interested but am not sure whether I would have the time and energy to devote to learning it at the moment, to be honest, with all the other things going in my life but I certainly hope that you have a high take up of interest!
Having just looked at the indepth chart via the link I would have to say I was quite astonished at how far-out the phonetic representation was... compared with that of the the natives: the one above! There were many ocacasions when the rise of the curve is not in any way matched by the phonetic representation below
Lazarus in view of that observation how then can such a system possibly help people learning languages when it appears so inaccurate? I would be interested in knowing the answer to this mystery/puzzle.
I think it was used in our French language texts published in Australia around 1970. I always found it confusing. One thing I have noticed about it is that if you look at a simple word like "cat", you will find a couple of different pronunciations when the actual number is probably over 10.
This is the first I have heard of it, but it is fascinating.
I would like to learn more.
I have never seen this system before, but I think it is awesome. I like that it shows the different sounds of the same letters. I definitely don't think it's useless if you can say words you haven't even heard before in a language that you never have studied.
