The Shadowing Technique: How to Use It to Speak a Language
The shadowing technique means listening to a native speaker and speaking along with them almost at the same moment, a fraction of a second behind, copying their sounds, rhythm and melody. Done for ten to fifteen minutes a day with short audio you already understand, it is one of the most direct ways to make your speech sound more natural and to get your mouth used to saying whole sentences at normal speed.
This guide goes deeper than the quick version in our post on how to improve your speaking skills: a full routine, which audio to pick, the mistakes that waste the time, and how to turn a conversation partner into your best shadowing source.
What is the shadowing technique?
Shadowing comes from the training of simultaneous interpreters, who practise repeating a speaker’s words in real time before they learn to translate them in real time. The polyglot Alexander Arguelles made it popular with self-learners, often shadowing while walking outdoors to keep his energy and volume up.
In language teaching, the best-known academic treatment is Shuhei Kadota’s book Shadowing as a Practice in Second Language Acquisition (Routledge, 2019). Kadota describes shadowing as helping in several ways at once: it sharpens listening, it gives your short-term memory practice holding and rehearsing new words and phrases, it rehearses parts of the process of producing speech, and it trains you to monitor your own output.
The key difference from ordinary “listen and repeat” is timing. With listen and repeat, you wait for the sentence to end and then say it. With shadowing, you start speaking while the speaker is still talking. That leaves no time to translate or think about grammar, so you copy the sound as a whole: where the stress falls, where the voice rises and drops, which words get swallowed.
Why shadowing works, and what it won’t do
Most learners who have studied for a while can read far better than they can speak. Part of the gap is physical. Your mouth has never produced these sound sequences at speed, and your ear has not learned which parts of a sentence matter. Shadowing attacks both at once, with hundreds of repetitions of real, natural sentences in a single session.
It is also low-pressure. There is no one waiting for your answer and no one to judge you, which is why it suits people who find talking to strangers draining (our introvert’s guide has more on that).
- Improves rhythm, stress and intonation, which shape an accent more than single sounds
- Gets whole phrases into your mouth at native speed
- Sharpens listening, because you have to hear every syllable to copy it
- Needs nothing but audio, a transcript and ten minutes
- Does not teach you to form your own sentences or answer questions
- Easy to do on autopilot without really listening
- Frustrating with audio that is too fast or too hard
- Gives no feedback unless you record yourself or ask someone
That last row of limits matters. Shadowing is a pronunciation and fluency drill. You are borrowing someone else’s words, so it cannot replace the moment where you have to find your own. Treat it as warm-up and polish for conversation, not a substitute for it.
The main types of shadowing
People use the word for several slightly different exercises. You will usually move through them in this order with each new clip.
| Type | What you do | Best for |
|---|---|---|
| Mumble (silent) shadowing | Move your lips or whisper along with the audio | First pass, catching the rhythm |
| Script shadowing | Shadow while reading the transcript | Getting every word right |
| Blind shadowing | Shadow with the transcript hidden | Real listening and real speed |
| Chorusing | Speak exactly with the audio, not behind it | Matching melody precisely |
| Delayed repetition | Pause after each sentence and repeat it | Beginners and hard material |
If full shadowing feels impossible at first, start with delayed repetition and work toward speaking over the audio. That is normal, not a sign that the method isn’t for you.
How to shadow: a step-by-step routine
Here is a routine that takes ten to fifteen minutes a day. You use the same short clip for several days, then move on.
-
Pick a one to two minute clip with a transcript
Something at or slightly above your level, spoken by one or two people at a natural speed.
-
Listen once without the text
Just follow the meaning. Notice how much you catch.
-
Read the transcript and look up what you don't know
You should understand everything before you start copying it.
-
Mumble along with the audio
Whisper or move your lips, following the rhythm rather than the words.
-
Shadow with the script
Speak out loud, a split second behind the speaker. Exaggerate the melody.
-
Shadow without the script
Hide the text. If you lose the thread, stop, replay that sentence, and continue.
-
Record yourself and compare
Play your recording next to the original. Pick one or two differences to fix tomorrow.
A few details make a big difference:
- Loop, don’t cover ground. Ten minutes on the same ninety seconds beats ten minutes on a whole episode. The value is in the repetitions.
- Stay with a clip for three to five days. Move on when you can shadow it blind without stumbling, or when you are bored.
- Speak at full volume. Quiet shadowing trains quiet, hesitant speech. Arguelles shadowed outside for a reason.
- Slow the audio if you must. Most players let you drop to 0.75x. Go back to normal speed as soon as you can, because natural speed is the point.
What materials should you use for shadowing?
The right material is short, clear, natural, and comes with a transcript. It should also be something you would actually want to say. Shadowing a news anchor gives you a news anchor’s voice, which sounds odd in a café.
| Material | Level | Why it works | Watch out for |
|---|---|---|---|
| Learner podcasts with transcripts | Beginner to intermediate | Clear speech, transcripts included | Can be slower and more careful than real speech |
| Street interviews with subtitles (e.g. the Easy Languages channels on YouTube) | Intermediate | Real people, real speed, everyday topics | Background noise, many speakers |
| Audiobook plus the ebook | Intermediate to advanced | Long, clean, professional narration | Narration style, not conversation |
| Dialogue from shows or films with subtitles | Intermediate to advanced | Emotional, natural phrasing | Subtitles often differ from what is said |
| Textbook audio | Beginner | Matches what you are studying | Often stiff and unnatural |
| Voice messages from a native speaker | Any | Everyday language, a voice you know, you can ask for more | Needs a partner |
Two rules for choosing:
- Understand at least most of it before you start. Shadowing is not the place to learn new material from scratch. Understanding comes first, as we explain in our guide to comprehensible input.
- Pick one speaker you like. Copying the same voice for a few weeks builds a consistent accent faster than jumping between regions.
Common shadowing mistakes
Most people who try shadowing and give up made one of these mistakes.
Using audio that is too hard. If you have to stop every few seconds, you are not shadowing, you are deciphering. Drop a level.
Using clips that are too long. A twenty-minute episode means you hear each sentence once. Cut it down to a minute or two.
Copying the words but not the music. Most of what makes speech sound foreign is rhythm, stress and intonation rather than individual sounds. Exaggerate the rises and falls, the pauses, the linked words.
Never recording yourself. You can’t hear yourself accurately while speaking. A thirty-second recording each session shows you what to fix.
Shadowing on autopilot. After a few days you can mouth a clip without thinking about its meaning. That is the moment to move on or to switch to blind shadowing.
Skipping the step where you use the phrases. A phrase you have shadowed fifty times is ready to use. If you never say it to a person, it stays a party trick. Put the best phrases on flashcards (see our spaced repetition guide) and use them in your next conversation.
Combine shadowing with a language partner
The weakness of shadowing is that it has no feedback and no conversation. A native speaker fixes both, and a text-and-voice language exchange is a convenient way to get one. If you have never done an exchange, start with how to find a partner and how to run an exchange.
Here is a simple loop that works in any chat app with voice messages, and which we built LangX around:
- Write a few sentences about your day in the language you are learning and send them to your partner.
- Get them corrected. In LangX you hold a message to correct it, and corrections are unlimited on every plan, so your partner can fix every line.
- Ask your partner to send the corrected text as a voice message. Now you have native audio of your own sentences, which is the most useful shadowing material there is, because it is exactly what you want to say.
- Shadow it, then send back your own voice message and ask what still sounds off.
When your partner isn’t around, you can also hold any message in a LangX chat to hear it read aloud by a synthetic voice. That is handy for checking how a written message sounds, though a real partner’s recording is better for shadowing. Read-aloud has a daily allowance that depends on your plan.
LangX matches you only with people who speak the language you are learning and are learning yours, so helping you with a recording is a fair trade: you do the same for them. The free plan lets you start 5 new conversations a day, with unlimited replies and corrections. LangX has voice and photo messages but no live calls, so if you want to practise speaking on a live call, a tutor or a meetup fills that gap.
A sample week of shadowing
If you want a starting plan, this fits into fifteen minutes a day:
- Days 1 to 3: one clip from a learner podcast. Full routine every day, recording on day 3.
- Days 4 and 5: a voice message from a partner, based on your own corrected sentences. Shadow it, send yours back.
- Days 6 and 7: a new clip, slightly harder, such as a street interview. Compare your day 7 recording with day 1.
Once the habit is in place, you can fold shadowing into a wider speaking routine, like the 30-day plan in our speaking guide, and use the phrases you learn to start thinking in the language.
FAQ
Does shadowing really work?
It works well for what it trains: pronunciation, rhythm, intonation and the speed at which you can produce familiar phrases. Researchers such as Shuhei Kadota have studied it in classrooms, mostly with learners of English. It does not teach you to build new sentences on your own, so combine it with conversation.
How long should I shadow each day?
Ten to fifteen minutes is enough for most people. Short daily sessions work better than an occasional hour, because the goal is many repetitions of a small amount of audio.
Can beginners use the shadowing technique?
Yes, with easy material. Start with slow, clear audio such as textbook dialogues or beginner podcasts, and use delayed repetition (pause, then repeat) until you can keep up with the speaker.
Should I shadow with or without the transcript?
Both, in that order. Shadow with the script first so you get every word right, then hide it so you are really listening.
What is the difference between shadowing and repeating?
When you repeat, you wait until the speaker finishes and then say the sentence. When you shadow, you speak while they are still talking, slightly behind them, which forces you to copy the sound directly instead of reconstructing it.