Is Your Shadowing Actually Working? A 5-Sentence Self-Test

Shadowing French for weeks and not sure it's helping? Five sentences, three recordings, ten minutes. Find out which part of your pronunciation shadowing is fixing, and which part it can't reach.

S

Simply French Team

11 min read

You have been shadowing for a few weeks. Every day you play a clip, speak over it half a beat behind, and match the speaker as closely as you can. It feels productive. It even sounds good, while the audio is playing.

The question nobody answers is whether it is working. Shadowing does not come with a scoreboard. You finish a session feeling closer to the model speaker, but "feeling closer" is exactly the judgement your own ear is worst at making.

This page gives you a way to check. It takes about ten minutes: five French sentences, three recordings each, and a checklist. By the end you will know which part of your pronunciation shadowing is improving, and which part it structurally cannot reach.

What shadowing trains, and what it can't

Shadowing is good at one thing, and it is a big thing: the shape of the sentence. Rhythm, the rise and fall of intonation, where the pauses go, how words run into each other. These are large, slow-moving features. You can hear them in the model, you can hear them in yourself, and when the two disagree the mismatch is obvious.

It is much weaker at the small things: individual sounds. Whether your tu is really /y/ or slides into tout. Whether your on is a nasal vowel or a vowel plus an English n. Those differences are a fraction of a second long, and while you are shadowing they are covered by the model speaker's voice in your ears.

There is a second, deeper problem. Your ear sorts sounds into the categories your first language taught it. If English has no /y/, your ear tends to file a French /y/ and your own slightly-wrong attempt into the same box. So you can play back your recording, listen carefully, and honestly hear no difference, even when a French listener would.

So the test below is built to separate the two. Some checks you can grade yourself. Some you cannot, and the test says so.

The setup: three recordings per sentence

For each of the five sentences, you will make three recordings on your phone:

  1. Model. Find a native recording of the sentence, or have a speaker or a text-to-speech voice you trust say it. This is the reference.
  2. Shadowed. Play the model and speak over it, the way you normally shadow. Record only your voice (headphones help).
  3. Cold. No audio. Say the sentence from memory at natural speed, once, without a second try.

The third recording is the one that matters. Shadowing is easy with the model in your ear: you are borrowing its rhythm in real time. The question is what is left when the model is gone. If your cold recording sounds like your shadowed one, the shadowing has transferred. If it falls back to your old habits, it hasn't yet.

The 5 sentences

Each sentence targets one thing shadowing is supposed to fix. IPA uses · for syllable breaks, which is how the sentence is actually chunked in speech, not where the word boundaries are.

1. Linking: Il est arrivé avec une amie.

SentenceSpoken asMeaning
Il est arrivé avec une amie.[i·lɛ·ta·ʁi·ve·a·vɛ·ky·na·mi]He arrived with a friend.

Look at the syllables: il est becomes [i·lɛ], est arrivé becomes [ta·ʁi], avec une becomes [vɛ·ky], une amie becomes [na·mi]. Almost every word hands its last consonant to the next one. This is enchaînement plus liaison, and it is the thing shadowing trains best.

Check (you can grade this yourself): in your cold recording, is there any audible gap or glottal stop between words? Listen especially at est‿arrivé and une‿amie. A gap means you are still speaking word by word.

2. Dropped sounds: Je ne sais pas ce que tu veux.

SentenceCarefulCasualMeaning
Je ne sais pas ce que tu veux.[ʒə·nə·sɛ·pa·sə·kə·ty·vø][ʃɛ·pa·skə·ty·vø]I don't know what you want.

In everyday speech, je ne sais pas collapses to chais pas, and ce que shrinks to [skə]. If your model is casual and natural, it will sound like the right-hand column. Most learners shadow it correctly and then say the left-hand column cold, every syllable intact, because that is how the sentence looks on the page.

Check (you can grade this yourself): count the syllables in your cold recording. Eight means you are reading, not speaking. Five or six means the casual form has stuck.

3. The vowel your ear can't grade: Tout le monde est venu, sauf vous.

SentenceSpoken asMeaning
Tout le monde est venu, sauf vous.[tu·lə·mɔ̃·dɛ·və·ny·sof·vu]Everybody came, except you.

Three ou sounds ([u] in tout, vous) and one u sound ([y] in venu). For an English speaker, venu is the hard one: it tends to come out as venou, which sounds like a different word.

Check (you probably can't grade this yourself): play back your cold recording and ask whether the vowel in venu is clearly different from the vowel in vous. If you hear a difference, good, but be suspicious of a "yes". This is the contrast where a learner's ear and a French listener's ear disagree most often. If you can't tell, that is the honest answer, and it is the point of this sentence.

4. Nasal vowels: On en a parlé pendant longtemps.

SentenceSpoken asMeaning
On en a parlé pendant longtemps.[ɔ̃·nɑ̃·na·paʁ·le·pɑ̃·dɑ̃·lɔ̃·tɑ̃]We talked about it for a long time.

Five nasal vowels in seven words, and two kinds: [ɔ̃] (on, long) and [ɑ̃] (en, pendant, temps). The usual mistakes are pronouncing an n or ng after the vowel (longtemps with a full English ng), and merging [ɔ̃] and [ɑ̃] into one sound.

Check (half and half): the extra consonant is easy to hear: listen for an n or ng at the end of pendant and longtemps. Whether your [ɔ̃] and [ɑ̃] are actually different is the same problem as sentence 3: your ear may hear them as one.

5. Rhythm and the question rise: Tu viens ce soir, ou pas ?

SentenceSpoken asMeaning
Tu viens ce soir, ou pas ?[ty·vjɛ̃·sə·swaʁ·u·pa] ↗Are you coming tonight, or not?

French puts its weight on the last syllable of each group, not on particular words: tu viens ce SOIR, ou PAS ? And a yes/no question asked without est-ce que is carried almost entirely by the pitch rising at the end. English speakers tend to stress VIENS and let the end fall, which makes the question sound like a statement.

Check (you can grade this yourself): in your cold recording, is soir longer than viens? Does your voice go up on pas? Hum the sentence with your mouth closed if it helps. The melody is easier to hear without the words.

Score yourself

Give yourself one point per sentence where the cold recording passes the check:

SentenceWhat it testsSelf-gradable?Point
1Linking, no gaps between wordsYes☐
2Dropped syllables in casual speechYes☐
3/y/ vs /u/No, not reliably☐
4Nasal vowels, no added nPartly☐
5Final-syllable stress, question riseYes☐

What the result means:

  • Shadowed passes, cold fails, on 1, 2 and 5. Shadowing is working during the session but not transferring. The fix is not more shadowing. It's adding the cold step to every session: shadow three times, then say it once alone.
  • Cold passes on 1, 2 and 5. Shadowing has done its job on the shape of your French. Your rhythm, linking and melody are moving in the right direction. Keep going.
  • You marked 3 and 4 as passes. Maybe. But those are the two checks this test cannot verify, and a pass you graded yourself on a vowel contrast is the least trustworthy number on this page.

Most people who take this honestly land in the same place: solid on rhythm, unsure on sounds. That is not a failure of effort. It is what shadowing alone produces.

Why shadowing can't fix sounds on its own

When you shadow, you are copying what you perceive. If your perception already merges /y/ into /u/, you copy the merged version, and it sounds right to you because it matches what you heard. Repetition makes the habit more fluent, not more accurate.

That is the gap in shadowing as a method. It has an input (the model) and an output (your voice), but no comparison step that works below the level your ear can resolve. Everything above that level (rhythm, linking, melody) gets corrected naturally, because you can hear the mismatch. Everything below it doesn't, because you can't.

So the useful question is not "should I keep shadowing?" (yes, for rhythm it is one of the best things you can do). It is: what checks the sounds? A teacher can, if you have one. A French friend can, if they're willing to be blunt. Or something that compares your sound to the target sound directly, without going through your ear first.

Frequently asked questions

How long should I shadow before taking this test?

Two to three weeks of regular practice is enough for rhythm and linking to show up in a cold recording. Take the test, keep the recordings, and take it again a month later. Comparing your own cold recordings across a month is the most honest progress check you have.

Should I shadow at full speed?

Start slower if you need to, but the cold test should be at natural speed. Most of what makes French sound French (linking, dropped syllables, final-syllable weight) only happens at speaking speed. A slow, careful version of sentence 2 will always have eight syllables.

My shadowed recording sounds great. Why does my cold one fall apart?

Because the shadowed one wasn't fully yours. With the model in your ears you are borrowing its timing in real time. That is useful practice, but it measures how well you can follow, not what you can produce. The cold recording is the one other people will hear.

Can I use any audio for shadowing?

Use audio where you can find or make a transcript, and where the speaker talks at natural, conversational speed. Audiobooks and news are often too careful: they keep every ne and every e, so you train the left-hand column of sentence 2 instead of the right.

Is this test enough to know my accent is good?

No. It tells you whether shadowing is working. For a broader check of your accent, the self-test for your French accent covers more ground.

Say them and get scored

Sentences 1, 2 and 5 you can grade with your own ears. Sentences 3 and 4 you can't. Not because you aren't listening carefully, but because the check has to happen below the level your ear sorts sounds at.

Say the five sentences into Simply French and each one is scored sound by sound: the /y/ in venu, the nasal in longtemps, the linking at une‿amie. Do it once now, keep shadowing, and do it again in a month. That is the scoreboard shadowing never came with.

Related

Tags

#Learn French#French pronunciation#French speaking
Free for 7 days

Now find out how you actually sound.

Simply French listens to you say each sentence and scores it word by word, so you can hear the difference you have just read about — and fix it while it is still fresh.

Takes about 10 minutes. No card needed to start.