F, V, Z for Korean Speakers: Why 'Fan' Sounds Like 'Pan'

A Korean colleague at lunch says "I'm a big paen of berry vivid mubies" and means "I'm a big fan of very vivid movies." None of those substitutions are random. Korean's phoneme inventory has no /f/, /v/, or /z/, so when an English word asks for one, your mouth hands over the closest hangul it does have: ㅍ for /f/, ㅂ for /v/, ㅈ for /z/. The substitutions are predictable, the fix is mechanical, and the daily Konglish you're surrounded by (커피, 비자, 비디오) is quietly reinforcing the wrong sound every time you order coffee.

Three sounds. One missing mouth shape. A reflex that thirty years of Korean media has trained into your ear. Once you see the pattern, you can fix it in a couple of weeks.

Why your mouth doesn't have these sounds yet

Open a Korean phonology table and look for /f/, /v/, or /z/. They're not there. The Wikipedia Korean phonology reference lists ㅍ as /pʰ/ (an aspirated bilabial stop), ㅂ as /p//b/ (a plain bilabial stop), and ㅈ as /tɕ//dʑ/ (a palatal affricate). The native fricative inventory is /s/, /s͈/, and /h/. That's the entire list.

The problem isn't that your mouth can't make these sounds. It's that English /f/ and /v/ are labiodental fricatives: top teeth pressed lightly against the bottom lip with air leaking through. Korean has nothing in that position. ㅍ in 파도 (pado, "wave") starts with both lips sealed and pops open; /f/ in father starts with teeth on lip and never closes. To a Korean ear, the two sounds are similar enough that the brain swaps one for the other. To an English ear, they aren't.

/z/ has a different problem. It's an alveolar fricative: same tongue position as /s/, but with the vocal cords buzzing. Korean has /s/. It does not have a voiced fricative anywhere in its inventory. Voicing simply isn't a feature your phoneme system uses for fricatives. So when English asks for /z/, your mouth substitutes the closest thing it does know, which is ㅈ.

The substitution map you didn't know you had

Fan sounds like pan. Very sounds like berry. Zip sounds like jip. Here is the full version of what your tongue is doing without permission:

• /f/ → ㅍ: fan sounds like pan, fine like pine, coffee like kopi, fresh fish like fresh pish.
• /v/ → ㅂ: very sounds like berry, vest like best, love like lub, vivid like bibid.
• /z/ → ㅈ: zip sounds like jip (or sip), zoo like joo, zebra like jebra, prize like price.

Speech pathologists call this "stopping of fricatives": replacing fricative sounds (where air keeps flowing) with stops or affricates (where air gets blocked and released). It isn't laziness. It's the only motor program your mouth has on file for the slot in question.

The most common follow-on error: peas and peace sound identical, because the /z/-vs-/s/ distinction collapses when your tongue treats both as ㅅ. Native English ears then have to guess from context, and they often guess wrong. Yax Accent's notes on Korean speakers flag this as one of the highest-frequency intelligibility issues.

Konglish is quietly training you wrong

Every time you order 커피 (keopi) in Seoul, you are practicing /pʰ/ in the slot where English wants /f/. You did not invent that substitution. Korean did, decades ago, and now it pipes the wrong sound into your ear daily.

Konglish is the layer of English-derived loanwords that Korean has adapted to its own phonology. The adaptations follow exactly the substitution map above:

• 커피 (keopi), coffee. The /f/ became ㅍ long before you tried to order one in English.
• 비디오 (bideo), video. /v/ → ㅂ.
• 비자 (bija), visa. /v/ → ㅂ; /z/ → ㅈ.
• 비타민 (bitamin), vitamin. /v/ → ㅂ.
• 패션 (paesyeon), fashion. /f/ → ㅍ.
• 핸드폰 (haendeu-pon), handphone. /f/ → ㅍ in the second half, plus an epenthetic /eu/ to break the consonant cluster.
• 지그재그 (jigeujaegeu), zigzag. Both /z/s became ㅈ.

90 Day Korean's overview makes the underlying rule explicit: Konglish words are pronounced according to their hangul spelling, not the English original. That rule is exactly what makes Konglish dangerous for an ESL learner. Repetition entrenches habit. After a lifetime in Korean, you have heard the wrong substitution tens of thousands of times before you ever attempt the English version.

This is also why telling a Korean speaker "just say /f/" rarely works. The brain already has the word file for coffee. It's the wrong file.

The mouth move you've never made

Stand in front of a mirror, put your top teeth lightly on your bottom lip, and push air out without moving anything. That's /f/.

Now turn on your voice and feel the buzz. Same lip-and-teeth position. That's /v/.

This is the move ㅍ doesn't ask for. ㅍ uses both lips; /f/ uses one lip and one row of teeth. If you watch yourself in the mirror saying fan and your lips meet without your top teeth touching, your mouth produced pan. Restart.

The throat-finger trick disambiguates /f/ from /v/. 7ESL describes it cleanly: place your fingers on your throat and say fff. No vibration. Now say vvv. Buzzing. Same mouth shape, different vocal-cord setting. If you can feel the difference between the two, you've already done the hardest part: staying on lip-and-teeth, not retreating to ㅂ when the buzz turns on.

A quick test you can run on yourself today. Say very fine fan. If your mirror shows your lips pressing together at any point, the substitution is still active.

/z/ is the hardest one, and it's hiding in every plural

/z/ asks your mouth to do two things at once: push air through a narrow channel for friction, and run the vocal cords. Korean's only fricative in that position is /s/, which is unvoiced. Going from /s/ to /z/ feels like trying to whistle and hum at the same time the first few times you try it.

Here's why it matters more than you think. English plural -s is not always pronounced /s/. The rule, laid out by Learn English Sounds:

• After voiceless consonants, plural -s is /s/: cats /kæts/, cups /kʌps/, books /bʊks/.
• After voiced consonants and vowels, plural -s is /z/: dogs /dɒɡz/, bags /bæɡz/, days /deɪz/.
• After sibilants (/s, z, ʃ, ʒ, tʃ, dʒ/), plural -s is /ɪz/: horses /ˈhɔːrsɪz/, boxes /ˈbɒksɪz/.

The same rule applies to third-person singular verbs (he runs /rʌnz/, she works /wɜːks/) and possessives (Tom's /tɒmz/, Mike's /maɪks/).

Korean speakers default to /s/ for all three contexts, which makes dogs sound like /dɒɡs/ and Tom's car sound like /tɒms kɑːr/. Native English ears parse this as a small but persistent off-ness, like a slightly out-of-tune note. Individually each word might pass; in fast speech, the accumulated mismatch is what marks the accent.

Drill the contrast directly with minimal pairs: zip / sip, zoo / Sue, prize / price, peas / peace, raise / race. If you say each pair and they sound identical to you, you haven't made /z/ yet. You've made two /s/s.

The five-minute drill

Pick a mirror. Set a timer for five minutes. Run these in sequence:

1. /v/ line: very vivid violins. Say it ten times. Watch your bottom lip the entire time. If it ever leaves your top teeth, restart.
2. /f/ line: fast fish fingers. Ten reps. Clip the final consonants. Fish, not fishi. (Korean speakers tend to add a small /i/ to consonant-final words. Ignore that battle for now and focus on the /f/.)
3. /z/ line: lazy zebras zoom. Ten reps. Sustain the buzz on every /z/. If it sounds like lacy seebras soom, you devoiced. Start over.
4. Plural /z/ line: dogs, bags, days, dreams, friends. Ten reps. End each word on a clean voiced /z/.

Five minutes a day. Two weeks. The drill is what speech coaches consistently recommend. It isn't a clinical-trial result, but it's the cheapest, most-reproducible piece of practice available, and it works because it forces the same motor pattern hundreds of times per session.

For longer reps, the same drill scales up into actual conversation. An AI conversation partner like Conversa lets you run very vivid violins and dogs and cats across full sentences a hundred times in a row, at your pace, without anyone getting bored. The narrowing is what makes the practice stick: pick three sounds, ignore everything else.

What changes when you crack it

Once your mouth can produce /f/ vs /v/ and /s/ vs /z/, your ears start hearing things that were previously identical. Have to and had to stop sounding like the same word: native speakers reduce have to to /hæftə/ and had to to something closer to /hætə/, two clearly different shapes once you can hear voicing. Buzz and bus split into two words. The /z/ at the end of those, these, has, was, is becomes audible, and English connected speech stops feeling like a single river of sound. It starts breaking into separate words again.

The substitutions are not your fault. They're what your phoneme inventory does on autopilot, reinforced by every Konglish loanword you've ever heard. Once you can name the saboteur (your phoneme inventory), the gesture (top teeth on bottom lip, vocal cords on or off), and the rule (plural -s after voiced consonants is /z/), the fuzzy "voicing problem" becomes three habits you can fix. (Sister posts in the same series: English TH sounds for French speakers; English R for Brazilian speakers; on the Korean side, Korean batchim and liaison walks the reverse direction.)