Why Can’t I Hear French?

If you have studied French for years and still cannot follow two French people in conversation, the problem is almost never vocabulary, grammar, or effort. The problem is that your brain stopped registering several French vowel sounds when you were a child, and nobody has trained your ear to register them again.

I am Bernard Henusse. I have been teaching spoken French one-on-one to English-speaking adults in our home in Brittany since 2008, with around 400 alumni from more than thirty countries. The single most consistent thing I see in students who walk through our door is a perceptual gap that no app, grammar book, or group class has ever fixed for them. It has a name, it has a fix, and almost nobody in the language industry talks about it in plain language.

Why can’t I understand spoken French even after years of study?

Because spoken French uses sound distinctions that your English-trained brain learned to ignore before you turned eight, and no amount of vocabulary or grammar work will rebuild a perceptual system you cannot consciously feel is broken.

The pattern I see is always the same. A student arrives who has done five years of Duolingo, taken evening classes at the Alliance Française, paid for an italki tutor, possibly even moved to France and waited for immersion to do its job. They can read Le Monde, write a polite email, order in a restaurant. Then a French friend comes over for dinner, and inside two minutes they are catching every fifth word and the friend is switching to English.

The conclusion most of these students have already drawn is that they are bad at languages. They are not. What they are missing is not knowledge. It is perception, a physical skill that is trained by hearing yourself, being corrected, and producing the sound again, hundreds of times, with someone in the room.

What is categorical perception, and why does it stop adults from hearing French?

Categorical perception is the process by which your brain, between birth and roughly age seven, decides which sound distinctions are worth keeping and discards the rest. Every baby is born able to distinguish every sound in every language on earth. By the time the same child is eight, that universal capacity is gone. The brain has built an efficient filter that lets through only the sound categories that matter in the language spoken around it. Everything else gets routed to the nearest familiar category.

The standard reference is Patricia Kuhl at the University of Washington, who has spent decades documenting the process. Her central finding, supported by a large body of follow-up research, is that the perceptual map of your native language is essentially set early in life. Adults can still learn second languages, but the auditory system is no longer neutral. It forces foreign sounds into the closest English-shaped slot it can find.

For an English-speaking adult learning French this is serious. French uses a working set of about 14 distinct vowel qualities in normal speech, and at least eight of those are not in your English inventory at all. Without training, you hear them as approximations of vowels you already know.

When a French speaker says dessus (on top of) and dessous (underneath), your brain registers both words as containing the same vowel. They do not. To a French ear they are as different as pen and pain are to yours. When you produce them as the same sound, French speakers do not hear a small accent. They hear ambiguity, and their brain does the extra work of guessing from context. After thirty seconds of that, they are tired, and they switch to English.

Which French sounds is your brain most likely filtering out?

The eight French vowel distinctions English speakers most reliably miss fall into three groups: the three nasal vowels, the front-rounded vowels that have no English equivalent, and the open/close pairs of mid-vowels that English flattens into a single sound.

The three nasal vowels. French has three distinct nasalised vowels. English has none. You can hear them in un, on, and en. Most English-speaking learners flatten all three into a generic nasalised “on” sound. To a French ear, vin (wine) and vent (wind) are completely different vowels. The same is true of brun (brown) and brin (sprig).

The front-rounded vowels. French uses a set of vowels produced by rounding your lips as if for oo, while holding your tongue forward as if for ee. The /y/ in tu is the famous example. Your ear, never having had to handle this combination in English, files it under oo and produces tu as tout. The same problem hits the /ø/ in deux and the /œ/ in peur. Compare deux with de, or peu with pu. If your ear is trained, you hear two different words. If not, you do not.

The open/close mid-vowel pairs. French distinguishes a closed é from an open è, and a closed o from an open ô. Pêcheur (fisherman) and pécheur (sinner) differ in exactly this way. So do saute (jumps) and sotte (silly). On paper it looks pedantic. In speech it changes the meaning of the word.

None of these vowels is rare. They appear in everyday words. The gap is not in literary French. It is in the French your neighbours are speaking across the dinner table.

Can apps and tutors fix this on their own?

No. Apps, group classes, online tutors, immersion, and pronunciation videos all have a real function. None of them is built to retrain categorical perception in an adult ear. The reason is mechanical, not motivational.

Apps like Duolingo and Babbel build vocabulary and reading comprehension. They cannot train your ear because they cannot hear you. When the app plays dessus and dessous, it has no way to know whether you perceive them as two sounds or one. You tap the right button, the app says correct, and your perception has not moved.

Group classes at the Alliance Française or your local language school are good for community and motivation. In a class of ten students for sixty minutes, you get roughly five minutes of personal speaking time. Your teacher cannot stop the class for twenty minutes to work on the way your tongue fails to reach the front-rounded position for tu. They have nine other students. Five minutes a week is around 1% of what an adult perceptual system needs.

Online tutors on italki or Preply are better, because the attention is one-on-one. But a Zoom tutor for an hour a week is usually covering conversation, grammar, vocabulary, and pronunciation in the same session, and the phonetic work that retrains an ear needs sustained, focused, physical effort. Drilling minimal pairs for hours, not minutes. Watching your mouth in real space. This is the one part of language teaching that is genuinely hard to do through a screen.

Living in France is the one that breaks expats’ hearts. The assumption is that immersion will do the work. Instead they have been in the country for three, five, sometimes ten years, and still cannot follow a casual dinner conversation. Exposure without correction does not rewire perception. If your brain currently categorises dessus and dessous as the same vowel, hearing them five hundred more times will not change that. You need someone to stop you, point at the distinction, correct what comes out, and do that for hours.

Pronunciation videos are useful for theory. You can watch every phonetics video on the internet and still not have a trained ear, because hearing what you produce and having a trained listener correct it in real time is not something a video can do. It is the difference between reading a book about tennis and playing a match.

I am not saying these tools are bad. If a student cannot travel, I tell them Geri Metz’s online phonetics course at pronouncingfrench.com is excellent theoretical groundwork. The point is that none of these tools alone will retrain an adult ear. The job requires three things at once: a teacher trained in French phonetics, sustained intensity over consecutive days, and an environment where you produce French constantly. That combination is rare.

What does ear training actually look like?

Ear training is a physical, perceptual skill. It looks more like learning an instrument than learning a subject. You sit across from a teacher, you listen, you produce a sound, the teacher corrects what came out, and you produce it again, hundreds of times, until your perception has shifted.

The work has three layers. The first is the vowel inventory itself. My son Theo or I sit across from a student and drill the distinctions their ear has been filtering out. Lip position for tu versus tout. Tongue position for peu versus pu. The opening of the mouth for pêcheur versus pécheur. The student listens, produces, gets corrected, produces again. By the end of the first day, distinctions that were inaudible on arrival are audible. We record students on the first morning and the third afternoon, so they leave with proof of the shift. This is what we call ear training.

The second layer is rhythm and melody. French has a different musical shape than English. Stress is even, not heavy on one syllable. Phrases group into rhythmic units the French ear segments automatically. Getting rhythm wrong is one of the main reasons your French sounds laboured even when your grammar is fine. Bradley Cooper is famous in France for speaking excellent French even though his vocabulary is basic, because his rhythm is right.

The third layer is producing French from sound, not from text. Most of what has gone wrong in an adult learner’s French is that the brain has been silently reading written French behind the spoken stream, visualising the page while the speaker is talking, which guarantees you fall a sentence behind. The fix is hours of producing French out loud in response to French heard, with no text in front of you. These three layers, the sounds, the rhythm, and producing without reading, are the working definition of our method.

The format that makes this possible is one-on-one, in person, sustained over consecutive days. Not because the format is fancy, but because the work is mechanically incompatible with anything else. We run our 3-day programme, the 14 Sounds Experience, exactly this way, in our home in Brittany, from €1,900 all-inclusive (see pricing & packages). The constraints I just described are the constraints of the work, regardless of who delivers it.

Common questions

Am I too old to retrain my ear for French?

No. The plasticity that closed at age seven was the universal infant plasticity, which is a different thing from adult learning. Adults retrain perception slower than infants do, and the work has to be more deliberate, but the system is not locked. I have had students in their seventies who finished three days hearing distinctions they had been missing for fifty years. The constraint is the format of the work, not the age of the learner.

How long does it take to start hearing the missing sounds?

Most students start to register the distinctions inside the first day. They sit down at lunchtime on day one and say something like, oh, those are completely different sounds, how did I never hear that. The fuller change, where your ear processes the new categories automatically, takes longer and depends on practice after you go home.

Will training my ear also fix my French accent?

Largely, yes. Most of what English speakers think of as a French accent is downstream of perception. You produce vowels you can hear. If your ear cannot distinguish peu from pu, you cannot reliably produce them differently either. We do not promise students they will sound native. We do promise that French speakers will stop switching to English in the first thirty seconds.

Can I do the ear training online from home?

Partly. We offer follow-up Zoom sessions for students who have already done the in-person programme. The initial perceptual unlock, the part that depends on a teacher physically watching your mouth and correcting you in real time over consecutive hours, is genuinely hard to do through a screen. Online is good for maintenance once the ear has been opened. It is a poor format for the first opening.

If something here rang true

Book a free 15-minute call with me. No pitch, no commitment. I want to understand where your French has stalled and what would actually move it. If we are the right fit, I will say so. If you would be better served by another route, I will say that too.

Book a slot at /contact-us/, or write to me directly at bernard@realfrench.co. I read everything.

Bernard Henusse, Founder of Real French. Kerfiac, Brittany. Teaching spoken French through phonetic ear training since 2008.