Crash Course: International Phonetic Alphabet (IPA)

The IPA is a really cool part of linguistics. It's an alphabet that's meant to have every sound represented by one symbol, and every symbol has only one sound. That means that unlike English, you can always tell exactly how a word will sound when you read it in IPA. The alphabet is also expansive, accounting for sounds in other languages. If you know how to read the IPA, you can learn how a word should be pronounced in a foreign language without having to resort to weird descriptions like 'make the t softer'.

The practical benefits are clear in a number of fields. Opera singers use the IPA to sing in foreign languages without a heavy foreign accent. Speech pathologists use the IPA to identify mistakes that their patients make and work with them on them. Dialect coaches use the IPA to teach others how to successfully mimic a different dialect. Language documentation workers use the IPA to tell each other the sounds of a language that hasn't been recorded before. In other words, if you've ever wanted to compare speech sounds, you want the IPA.

Now the IPA is scary looking. The charts are big and they have a lot of unfamiliar symbols. A lot of these symbols don't represent the sounds you expect them to. There are a lot of new vocabulary words that make it a chore to study. I'm here to make the process of learning the IPA a little easier by providing a study guide you can use. For the most part, all you need is Wikipedia. Yes, it's that easy! You won't learn the IPA in a day - it takes a lot of practice to internalize it. But if you apply this approach piece by piece to the charts, you can tackle the thing slowly until you eventually get the whole thing. This is a formalization of my own ad hoc process when I taught myself the IPA back in high school. It may seem like a lot of steps, but if you follow it, you will definitely understand the IPA well, making it a powerful tool.


Let's start with vowels - there are fewer total vowels to learn, so the chart is less intimidating. The vowels are placed in a trapezium shape. For a full version of the trapezium with audio samples, check out the Wikipedia page.

You'll notice that the left side of the trapezium (the 'front') is smaller than the back. This is because the trapezium is supposed to represent the shape of the mouth. As you get closer and closer to your teeth, the space narrows and there is less space to go up and down. As you go back, there is more empty space to move the tongue around in. The below image shows the trapezium with a mouth over it.


Let's start by looking at the distinction between front and back. Go back to the Wikipedia page and click on one of the vowels near the front, like the [i]. Listen to how it sounds. Try to make the sound yourself. Notice the description of the vowel. If you speak any other languages, it may help to look at the 'examples of this sound in other languages' so that you can make a mental inventory of other words where this vowel appears. If the vowel does not appear in any language you know, listen to it anyway. Now do the same thing with a vowel in the back, like [u]. Make sure it's on the same line and it's the same order in the vowel pair. Listen to both vowels front to back. Perform both vowels. Really pay attention to how your tongue moves in-between each vowel. Try to smoothly transition from one vowel into another. This is the front-back distinction. Do this for each vowel height.

Now look at vowel height. Take a vowel that's high up and listen to it - let's use [i] again. Compare it with a slightly lower vowel, like [e]. Then compare it with an even lower vowel [ɛ], then go a little further to [æ]. Say each vowel aloud. Try to say the vowels one after another, in different sequences. Notice how your tongue is placed differently for each vowel. This is the low-high distinction (also known as the closed-open distinction).

The last one we'll be looking at is vowel roundedness. Look at one of the pairs - I recommend [i] and [y]. Listen to the one on the left - this is the unrounded version. Say the sound. Notice how your lips are positioned - are they spread thin or are they just in a neutral position? Now listen to the one on the right - this is the rounded version. Make the same sound you were just making, but this time make your lips into an 'o' shape. Notice how the sound changes as you bring your lips together. Transition back into an unrounded shape, then back into a rounded shape. Try to make the rounded shape first without having to do the unrounded shape. Listen again. This is the rounded-unrounded distinction.


There are some more distinctions to learn for consonants. You need to know this: (a) what is the place of articulation (where is your tongue when it's making the consonant?). (b) what is the manner of articulation (what is your tongue doing at that place?) and (c) is it voiced or unvoiced (are your vocal cords vibrating or not?) The place of articulation is represented at the very top of the chart, starting at the front of the mouth and moving all the way back in the mouth. The rows represent the manner of articulation. When a consonant pair appears, the one on the left is voiceless and the one on the right is voiced. For a full chart with audio examples, check out this Wikipedia page.

Place of articulation is basically anywhere in the mouth/lips. Starting the furthest out, you have bilabials. These are made using both lips. There are labiodentals, which is when your top teeth touch your bottom lip. Moving into the mouth, we have dentals. These are made with your tongue located behind your teeth. If you move your tongue back from there, you'll notice a little ridge - this is the alveolar ridge. Use the body of your tongue to feel the hard part of the top of your mouth - sounds made here are called palatals. Use the very back of your tongue at the edge between your hard palate and the soft part behind it. Sounds made here are called velars. (There are some more after this, but we'll stop here for now since no Standard English variety uses any place of articulation after this.).

Once your tongue has found the place it's going to be, the question is what is your tongue going to be doing. Consonants are made by impeding air flow somehow, so the manner of articulation is going to involve somehow messing with the air flow. A plosive, also known as a stop, is made by letting air pressure build up behind the tongue, and then suddenly moving your tongue down and letting the air escape. This makes a sort of 'explosive' sound, because you 'stopped' the air. Some examples of stops are 'p', 't', 'k', 'b', 'd', and 'g'. Perform each of these sounds, and notice how in each one, the air builds up a little behind your tongue before escaping.

A fricative is when you let the airflow go, but you use parts of your mouth to narrow the airflow. This results in some 'friction' and changes the sound. Some examples of fricatives are 'f', 'v', 's', 'z',  'sh', 'g' as in 'genre', 'th' as in 'thin,' and 'th' as in 'that'. Unlike stops, fricatives can be held for as long as you want. Perform each of the sounds listed before. Notice how the air is always moving. Try to extend the fricative for as long as you can.

An affricate is when you let air pressure build up, but instead of allowing it to escape suddenly, you let it escape in a controlled manner by disturbing the airflow. An affricate is therefore somewhat like a stop and a fricative together. Affricates don't appear on this chart because they are made of two sounds together, but they are important in describing sounds in English and other languages. Some examples of affricates are 'ch' ('t' + 'sh'), 'j' ('d' + 'g' as in 'genre'), and 'ts' ('t' + 's'). Perform some affricates. Then perform their component parts, one after the other. Notice how if you try to 'hold' a stop the same way you try to 'hold' a fricative, the result is an affricate.

A nasal is made by letting more air escape through the nose than through the mouth. An excellent way to know if a sound is nasal or not is by pinching your nose and then making the sound. If the sound is clear, then it's not a nasal. If your nose is blocked and you can't make the sound, it's a nasal. Nasal stops are just non-nasal stops with the air escaping through the nose. Say 'd' and then say 'n'. Notice how it's the same tongue movement, but the air is escaping through the nose. Moreover, because the air is escaping through the nose, you are not 'stopping' the air as much. This means you can hold a nasal the same way you can hold a fricative. Practice this with 'b' and 'm', and 'g' and 'ng' as in 'bang'.

Note there are some other, more complex manners of articulation that we won't be covering here. You've already got a lot to learn!

Finally, you need to know if the consonant is voiced or not voiced. Voicing refers to whether your vocal cords are vibrating. Your vocal cords vibrate when air passes through them - in other words, when you speak. Place your hands on your throat and say 'z'. Hold it. Feel your throat - it will be vibrating a little. Now place your hand on your throat and say 's'. Hold the 's' as long as you can. Feel your throat. Nothing is happening. Try to switch between the two. 'b' and 'p' are another example of a voiced/unvoiced pair. Try to get the hang for how it feels to 'add voice' or 'take voice away' from a sound.

Now that you understand what each of these concepts refers to, practice them by looking at sound names, reading them, and then trying to pronounce the sound. Go step-by-step. What is an 'unvoiced alveolar fricative'? Well right off the bat it's telling you it's unvoiced, so you know your vocal cords should not be vibrating when you make this sound. 'alveolar' is telling you that your tongue should be placed on that little ridge behind your teeth, so put your tongue there. 'fricative' is telling you that air should be passing between your tongue and the alveolar ridge, so let there be a small distance between your tongue and the alevolar ridge, and breathe out. If you did it right, you'll notice that the sound you just made is an 's'! Continue doing this for sounds you already know how to make until you can read a sound name and know what it means. Once you're comfortable doing that, you're ready to start reading the names of new sounds and trying to make those. The name of a sound is a recipe and each portion is an ingredient. Even if you've never made the dish before, you can make it by following the recipe. As you get better, you will remember how to do it by yourself and eventually you won't even need the recipe anymore.

Putting it all together

Now you know what the sounds are individually, you need to practice putting them together. Write some very simply sound sequences: 'p' + 'o' + 't' = pot. Read that aloud. You should notice that it sounds just like the word 'pot.' Try other simple sequences like this until you get comfortable seeing the symbols together and reading them. Get a little wild - make consonant clusters. 's' + 'p' + 'o' + 't' = spot. 'c' + 'a' + 'n' + 't' = can't. If you put a bunch of consonants together and you notice you have a hard time pronouncing them, go very slowly through each consonant. Practice holding stops and then transitioning to the next sound.

The final step is transcribing sounds. This step really sums up everything you've been doing up until now. A phonetics teacher once gave me some very good advice - 'write what you hear, not what you think you hear.' Put away what you know about your language and listen to the sound by itself. Listen to the sounds by themselves. Start simple - find a website that plays you random sounds and then transcribe them. Once you can do this for individual sounds, move on to sequences. Once you've done those, you can graduate to individual words. Say a word, slowly. Notice how your mouth moves while doing so. Listen to the word carefully. If it's an English word, ignore the spelling. Focus purely on the sound. Try to divide it into segments - how many different sounds can you hear? Then go through sequentially and work on each sound. Once you're finished, check your transcription with another source. As you get better, the entire process will get quicker until eventually it becomes entirely subconscious and you can turn a sound or a mental representation of sound into IPA transcription quickly. There's no special trick for sentences either - think of a sentence as a very long word. Sometimes someone will pause during a sentence, and you can note that, but otherwise you have no way of knowing what the divisions in a sentence are without knowing the language.

Presumably you are starting with a language you are familiar with. Move on to other languages. Practice the sounds you are not familiar with. Listen to them carefully. If you find there are some sounds you can't distinguish easily, look at the difference in their names (the 'recipe') and pay attention to what is different. Is it a rounded/unrounded distinction? Is it vowel height? Practice the distinction with other sounds, and then move on to the problematic sounds. Listen to the recordings, 'follow the recipe', and compare the result. You may not distinguish some sounds perfectly, and that's okay.

The rest comes with practice. There is no more advice I can give you beyond this point other than to practice, practice, practice. Be a mindful listener. If you're doing a mundane activity not requiring a lot of mental bandwidth, listen to what other people are saying - really listen to the sounds. You don't care about the meaning; you care about the waveforms reaching your ears. Practice transcribing mentally. Look at official transcriptions of words, and notice any mistakes you make consistently. Switch between a very broad transcription and transcriptions that are very specific. Learn about diacritics. The more you do it, the more you read about it, the more you see other people do it, and the more you hear it, the better you'll get. The great thing about practicing IPA is that once you've this past part down, you can practice it anytime, anyplace, anywhere.

1 comment:

  1. Very good course,well and clearly explained. Will let me understand fully this notion of fricate affricate and nasal, dental labiodental, alveolar,velar palatal and so one. Great thanks for allowing the free downloading. God bless you.