Shadowing English Speaking Practice: Step-by-Step Exercises

The idea is simple, and the results depend on how you practice: which audio you use, for how long and whether you record yourself. It helps most with rhythm and clarity, but it doesn't replace conversation. Combine it with English speaking practice at your level, where you have to say your own ideas instead of repeating someone else's.

What is the shadowing technique?

Shadowing is listening to English audio and repeating it out loud almost simultaneously, one or two seconds behind, without pausing to translate. Unlike "listen and repeat" sentence by sentence, there's no time to think: your mouth follows the speaker's rhythm, pauses and intonation.

It has been studied widely in Japan as a listening exercise (Hamada, 2016) and, more recently, as a pronunciation tool (Whitworth & Rose, 2025).

There are three basic ways to do it, from easiest to hardest:

• With a transcript: you read the text while you listen and repeat. It's the best way to start, especially at A1–A2.
• Without a transcript: you repeat by ear only. This forces you to actually listen instead of read.
• With a recording: you record yourself shadowing and compare your audio with the original. Several of the studies below used this version.

Does shadowing work? What the research says

The main reference is "A Systematic Review of Research on the use of Shadowing for Second Language Pronunciation Teaching" by Ben Whitworth and Heath Rose, published in 2025 in Research Synthesis in Applied Linguistics (vol. 1, no. 2, pp. 239–269; doi.org/10.1080/29984475.2025.2546827). The authors searched six databases and reviewed 44 studies. These are their conclusions:

• Overall, shadowing training can improve comprehensibility, intelligibility and accentedness, as well as aspects of pronunciation such as fluency and prosody (rhythm, stress and intonation).
• For individual sounds (vowels and consonants), results were inconclusive.
• Learners tend to see shadowing as interesting, enjoyable and effective.
• The studies have limits: many measure with controlled tasks (reading or repeating) rather than spontaneous speech, and few link acoustic measurements to real listener judgments.

Individual studies show how it is practiced and with what results:

Data from each study's published abstract (accessed Sep 29, 2026). Full links under "Sources".

One important point: the goal isn't to "sound native". A classic study by Munro and Derwing (1995) showed that a strong foreign accent does not necessarily reduce comprehensibility or intelligibility. So it makes sense to use shadowing to be understood more easily (rhythm, stress, pauses), not to erase your accent.

How to practice shadowing step by step

1. Choose a short audio at your level. 30 to 90 seconds, with a transcript and a clear voice. If you understand less than 70–80% without reading, it's too hard to start with (this figure is our practical guideline, not a research finding).
2. Listen 2–3 times without speaking. First for the general idea, then for details.
3. Read the transcript and mark it up. Underline stressed syllables, pauses (/) and words that link together in speech ("want to" → "wanna", "going to" → "gonna" in informal English).
4. Shadow with the text. Speak almost at the same time as the audio, slightly behind. If you get lost, don't stop: jump back in at the next sentence.
5. Shadow without the text. Once you've got it, repeat it by ear 2–3 times.
6. Record yourself and compare. Listen to your recording next to the original. Focus on one thing at a time: rhythm, intonation at the end of questions or -ed and -s endings.
7. Use it in a conversation. Pick 2–3 sentences from the audio and use them when you speak that same day. That turns what you repeated into something you can say yourself.

Shadowing routine and exercises by level (A1–C1)

As a starting point we use the frequency from Foote and McDonough (2017): at least 4 sessions a week of about 10 minutes. The table adapts material and focus to each CEFR level. If you don't know your level, see our guide to CEFR levels A1 to C2.

HelloVozi routine based on the frequency used by Foote and McDonough (2017). The times per level are a practical guideline, not a validated protocol.

A1 exercise: introduce yourself

Read and repeat this script (or ask someone to record it), stressing the syllables in capitals:

Hi, I'm ANna. I'm TWENty.
I'm a STUdent. I LIVE with my FAmily.
I LIKE MUsic and FOOTball.
NICE to MEET you!

Goal: don't separate every word. "Nice to meet you" is said almost as one word, with stress on "nice" and "meet".

A2 exercise: at the coffee shop

A: Hi! What can I get you? ↘
B: Can I have a coffee, please? ↗
A: Sure. Small or large? ↘
B: Small, please. How much is it? ↘
A: It's two fifty. ↘
B: Here you are. Thanks! ↘

Goal: intonation. In English, yes/no questions usually rise at the end (↗) and questions starting with "what" or "how" usually fall (↘). Exaggerate the arrows a little while shadowing.

B1 exercise: talk about a plan

I'm going to visit my cousins next weekend. / They live near the coast, / so we're planning to go to the beach on Saturday. / If it rains, / we'll probably stay home and watch a movie. / I haven't seen them for a year, / so I'm really looking forward to it.

Goal: pauses (/) and linking. Notice "going to" (in fast speech it sounds like "gonna"), "looking forward to it" (the words run together: looking‿forward‿to‿it) and that your voice doesn't stop until each "/".

B2 exercise: give an opinion

Personally, I think working from home has more advantages than disadvantages. For one thing, you save a lot of time on commuting. On the other hand, it can be harder to separate work from your private life. So in my view, the best option is a mix: two or three days at home and the rest in the office.

Goal: sentence stress. Stress the words that carry meaning (advantages, time, harder, mix) and keep grammar words (to, the, of) light. Linking phrases (for one thing, on the other hand, so in my view) are followed by a short pause.

C1 exercise: persuade

What we're proposing isn't a radical change. It's a simple adjustment that could save us a considerable amount of time. Now, I'm aware that some of you have concerns, and they're perfectly reasonable. But if we look at the figures from the last quarter, it becomes clear that the current system simply isn't working. So the real question isn't whether we should change — it's how quickly we can do it.

Goal: emphasis to persuade. Rehearse the contrast (isn't a radical change… It's a simple adjustment), the pause before But and the close (isn't whether… it's how quickly). Then retell the message in your own words in 30 seconds without looking.

A 4-week shadowing plan

The studies above lasted between 8 and 10 weeks. This 4-week plan is a realistic way to start and see whether it works for you. Keep the same audio level all month and change only the focus.

Plan suggested by HelloVozi. If it works for you, move on to the next audio level in month 2.

How to tell if you're improving

Progress is hard to notice while you practice. These three simple checks help you see it:

• Compare recordings. Record the same audio on day 1 and day 28. Listen to them back to back: are there fewer pauses mid-sentence? Does your question intonation sound more like the original?
• Ask someone to listen. What matters is being understood. Ask a friend or teacher to write down what you say in a 30-second recording. If they catch more words than before, your intelligibility has improved, which is exactly what several of the studies measure.
• Speak without a script. Once a week, talk for one minute about your day without preparing. Whitworth and Rose note that many studies only measure controlled tasks. Your real test is whether the rhythm you practiced shows up when you speak in your own words.

Where to find free audio for shadowing

• VOA Learning English: lessons organized by level (beginning, intermediate and advanced). Useful for B1 and for starting with news.
• BBC Learning English, 6 Minute English: roughly six-minute programs on many topics. Pick a 60–90 second extract for B1–B2.
• Your coursebook: the audio from the course you already use is perfect for A1–A2, because it comes with transcripts and the vocabulary you're learning.
• Series and videos with English subtitles (B2–C1): start with short scenes with a single speaker.
• Vary the accents once you feel comfortable. Alternate voices from the US, the UK, Australia and proficient non-native speakers. In real life you will hear English from many places, and shadowing different voices helps you recognize the same rhythm across accents. Keep one accent as your pronunciation model and use the others to train your ear.

Common shadowing mistakes

• Choosing audio that's too hard. If you don't understand the text, you're just making noise. Go down a level or use a shorter clip.
• Reading instead of listening. The transcript helps at first, but if you never drop it you're practicing reading aloud, not shadowing.
• Chasing perfection. Repeating the same sentence 20 times is tiring and doesn't add much. Do several full passes and move on.
• Never recording yourself. You don't hear yourself well while speaking. A recording is the easiest way to notice your real rhythm.
• Doing it once a week. The studies above used weeks of frequent practice, not one-off sessions. That's why short, frequent sessions work better.
• Stopping at repetition. Shadowing trains form, but speaking means saying your own ideas. That's the gap we explain in why understanding English isn't enough to speak it.

If you use ChatGPT, you can also ask it for scripts at your level to shadow. We share the prompts in ChatGPT English speaking practice: prompts that work. For guided missions, see Vozi Missions.

After shadowing, move on to speaking in your own words: these conversation topics by level include model texts to read aloud and questions to answer.

Next step: if you don't know your level, take the English level test (about 15 minutes, with instant results). To keep working on this topic, see how to practice English speaking with AI in Vozi Coach.