Languages
Japanese Shadowing Practice for Beginners: A Step-by-Step Guide
Shadowing means repeating Japanese audio almost at the same time as the speaker. It is one of the fastest ways to fix your rhythm and sound more natural, even as a beginner.
Many beginners can read hiragana and know basic phrases, but their Japanese still sounds "English". The words are right, but the timing, long vowels and intonation are not. Shadowing trains exactly these things.
This guide gives you a simple 5-step routine, a 7-day starter plan and three short N5-level passages with audio at slow and normal speed. Each passage has the full script in Japanese, kana, romaji and English.
When you finish the passages, you can keep going with real native speech in Corodomo's Japanese shadowing practice. Videos are split into single sentences, you can slow them down, and the AI gives feedback on your pronunciation and fluency.
- What is Japanese shadowing?
- Why shadowing works: rhythm, mora and intonation
- Before you start: kana and the right material
- The 5-step shadowing routine
- 3 beginner passages with audio
- A 7-day starter plan
- Common beginner mistakes and how to fix them
- FAQ
- Conclusion
What is Japanese shadowing?
In shadowing, you play a recording and speak along with it, about half a beat behind the speaker. You do not pause the audio. You follow the voice like a shadow follows a person.
This is different from "listen and repeat". In repeating, you wait until the sentence ends, then say it. In shadowing, you listen and speak at the same time. Your brain has no time to translate, so you copy the sound directly.
For beginners, shadowing works best with short, slow, clear audio. You do not need to understand fast anime dialogue on day one. Three or four simple sentences are enough.
| Method | What you do | Main benefit |
|---|---|---|
| Listen and repeat | Wait for the sentence to end, then say it | Accuracy, memory of sentences |
| Read aloud | Read the script with no audio | Reading fluency |
| Shadowing | Speak along with the audio, half a beat behind | Rhythm, speed, intonation, listening |
Why shadowing works: rhythm, mora and intonation
English is a stress-timed language. Some syllables are loud and long, others are short and swallowed. Japanese works differently. It is built on mora (拍, はく): small beats that each take roughly the same amount of time.
Most kana is one mora. But a long vowel, a small っ and ん each take a full beat too. English speakers often drop these beats because they do not exist in English. Shadowing makes you match the speaker beat by beat, so missing beats become obvious.
| Word | Mora split | Mora | Typical English-speaker version |
|---|---|---|---|
| 東京 (とうきょう, Tōkyō) | と・う・きょ・う | 4 | "TOE-kyo", 2 beats |
| 日本 (にほん, Nihon) | に・ほ・ん | 3 | "nee-HON", ん squeezed |
| 切手 (きって, kitte, stamp) | き・っ・て | 3 | "KEE-teh", no pause |
| 先生 (せんせい, sensei) | せ・ん・せ・い | 4 | "SEN-say", 2 beats |
| コーヒー (kōhī, coffee) | コ・ー・ヒ・ー | 4 | "KO-hee", long vowels cut |
Shadowing also trains intonation. You copy where the voice rises at the end of a question with か, and where it falls at the end of a statement. You also copy the small pauses after commas. These patterns are hard to learn from a textbook, but easy to absorb by copying a voice.
- Rhythm: every mora gets its own beat.
- Speed: you learn to say whole phrases without stopping between words.
- Listening: you start to hear long vowels, っ and ん in real speech.
- Intonation: you copy the melody of a sentence, not only the words.
Before you start: kana and the right material
The one real prerequisite is hiragana and katakana. You should be able to read kana slowly without romaji. Romaji hides the beats: "Tokyo" in English letters does not show the two long vowels. Kana shows them clearly: とうきょう.
You do not need kanji yet. Good beginner scripts show kana readings, or kanji with furigana. You also need to know what each sentence means before you shadow it. Shadowing sounds you do not understand turns into empty noise.
Choose material that is a little easier than your listening level. If you understand less than about 70% of a clip with the script, it is too hard for shadowing (this is a rule of thumb, not a fixed number).
| Your level | Good material | Length per session |
|---|---|---|
| Just finished kana | Textbook dialogues, slow self-introductions | 3–4 short sentences |
| JLPT N5 | Beginner podcasts, slow passages like the ones below | 20–40 seconds |
| JLPT N4 | Daily-life dialogues, NHK News Web Easy read slowly | 30–60 seconds |
| JLPT N3 and up | Vlogs, interviews, drama scenes, real YouTube videos | 1–2 minutes |
- Pick one voice you like. Copying the same speaker for a week makes progress easier to hear.
- Prefer natural speech over very dramatic voices. Anime characters often speak in exaggerated styles.
- Use audio you can slow down. A speed of 0.75x–0.85x is a good start.
The 5-step shadowing routine
This routine takes 10–15 minutes for one short passage. Do the steps in order. Do not skip to "shadow blind" on the first day. Each step prepares the next one.
- Listen (2–3 times, no script). Just listen. Try to catch any words you know. Notice where the speaker pauses and where the voice goes up or down.
- Read along (script + audio). Play the audio and follow the kana script with your eyes. Mark long vowels, small っ and ん with a pen. Check the meaning of every sentence.
- Shadow with the script. Play the slow clip and speak along, half a beat behind, while looking at the script. Do this 3–5 times. Do not stop the audio when you make a mistake.
- Shadow blind. Close the script. Shadow using only your ears. Start with the slow clip, then try normal speed. If you get lost, just mumble the sound until you catch up.
- Record and compare. Record yourself shadowing once. Listen to your recording and the model one after the other. Find one or two places that sound different and fix only those.
Step 5 is the step most learners skip, and it is the most useful one. While you speak, you cannot hear yourself well. A recording shows what really came out of your mouth.
On Corodomo's shadowing page you can do steps 3–5 sentence by sentence on real videos. You can replay one sentence, change the speed, and get an AI score for pronunciation accuracy and fluency, with unclear words highlighted. Full AI scoring is part of the Plus plan. Note that the AI does not grade pitch accent, so for pitch, compare by ear.
- Speak at a normal volume. Whispering hides mistakes.
- Focus on one thing per round: first rhythm, then long vowels, then intonation.
- Stop after 15 minutes. Short daily sessions work better than one long session per week.
3 beginner passages with audio
Here are three original N5-level passages. Each has a slow clip (0.85x) and a normal-speed clip. Use the slow clip for steps 1–4 and the normal clip when you can shadow the slow one without the script.
The audio uses a synthesized (TTS) voice. It is clear, but the intonation may be a little flatter than a real person's. Open the script only after you have listened 2–3 times.
Passage 1: Self-introduction
Focus: long vowels in とうきょう, べんきょう and どうぞ. The す in です and the し in します are almost silent, like a whisper.
Slow clip:
Normal clip:
Show transcript and translation
| # | Japanese | Kana | Romaji | English |
|---|---|---|---|---|
| 1 | はじめまして、マイクです。 | はじめまして、マイクです。 | Hajimemashite, Maiku desu. | Nice to meet you. I'm Mike. |
| 2 | アメリカから来ました。 | アメリカからきました。 | Amerika kara kimashita. | I'm from America. (lit. I came from America.) |
| 3 | 今、東京で日本語を勉強しています。 | いま、とうきょうでにほんごをべんきょうしています。 | Ima, Tōkyō de nihongo o benkyō shite imasu. | Now I'm studying Japanese in Tokyo. |
| 4 | どうぞよろしくお願いします。 | どうぞよろしくおねがいします。 | Dōzo yoroshiku onegai shimasu. | Nice to meet you. (a standard closing for introductions) |
Mora check: とうきょう = と・う・きょ・う (4), べんきょう = べ・ん・きょ・う (4), どうぞ = ど・う・ぞ (3).
Passage 2: My morning
Focus: the long vowels in コーヒー, and the small っ in さんじゅっぷん. Keep ん in でんしゃ as a full beat.
Slow clip:
Normal clip:
Show transcript and translation
| # | Japanese | Kana | Romaji | English |
|---|---|---|---|---|
| 1 | 毎朝六時半に起きます。 | まいあさろくじはんにおきます。 | Maiasa rokujihan ni okimasu. | I get up at 6:30 every morning. |
| 2 | 朝ご飯は、パンとコーヒーです。 | あさごはんは、パンとコーヒーです。 | Asagohan wa, pan to kōhī desu. | Breakfast is bread and coffee. |
| 3 | それから、電車で会社に行きます。 | それから、でんしゃでかいしゃにいきます。 | Sore kara, densha de kaisha ni ikimasu. | Then I go to work by train. |
| 4 | 会社まで三十分ぐらいかかります。 | かいしゃまでさんじゅっぷんぐらいかかります。 | Kaisha made sanjuppun gurai kakarimasu. | It takes about thirty minutes to get to work. |
Mora check: コーヒー = コ・ー・ヒ・ー (4), さんじゅっぷん = さ・ん・じゅ・っ・ぷ・ん (6). The っ is a short silent beat before ぷ.
Passage 3: A Sunday out
Focus: past tense ました, the long vowel in ケーキ, and the small っ in たのしかった. Let the voice fall gently at the end of each sentence.
Slow clip:
Normal clip:
Show transcript and translation
| # | Japanese | Kana | Romaji | English |
|---|---|---|---|---|
| 1 | 日曜日に、友だちと買い物に行きました。 | にちようびに、ともだちとかいものにいきました。 | Nichiyōbi ni, tomodachi to kaimono ni ikimashita. | On Sunday, I went shopping with a friend. |
| 2 | 駅の近くに、新しいカフェがありました。 | えきのちかくに、あたらしいカフェがありました。 | Eki no chikaku ni, atarashii kafe ga arimashita. | There was a new café near the station. |
| 3 | そこで、おいしいケーキを食べました。 | そこで、おいしいケーキをたべました。 | Soko de, oishii kēki o tabemashita. | There, I ate a delicious cake. |
| 4 | とても楽しかったです。 | とてもたのしかったです。 | Totemo tanoshikatta desu. | It was a lot of fun. |
Mora check: にちようび = に・ち・よ・う・び (5), ケーキ = ケ・ー・キ (3), たのしかった = た・の・し・か・っ・た (6).
A 7-day starter plan
This plan uses the three passages above. Each day takes about 15 minutes. The goal of week one is not perfect pronunciation. The goal is to build the habit and learn the 5 steps.
| Day | Passage | Steps | Target |
|---|---|---|---|
| 1 | Passage 1 | 1–3, slow clip | Follow the slow clip with the script, no stopping |
| 2 | Passage 1 | 3–5, slow then normal | Shadow blind at slow speed; record once |
| 3 | Passage 2 | 1–3, slow clip | Mark every long vowel, っ and ん in the script |
| 4 | Passage 2 + review 1 | 3–5 | Passage 1 at normal speed, no script |
| 5 | Passage 3 | 1–3, slow clip | Copy the falling intonation at the end of each sentence |
| 6 | Passage 3 + review 2 | 3–5 | Passage 2 at normal speed, no script |
| 7 | All three | 4–5, normal clip | Record all three; compare with your Day 2 recording |
On Day 7, keep your recordings. Listening to your first and last recordings side by side is the best way to see progress, and it keeps you motivated.
After the first week, try the Daily Challenge in the Corodomo community. There is a new short passage every day with native model audio. You read it aloud, get AI scores for pronunciation and fluency, and you have 3 tries per day. It is a simple way to keep shadowing daily without searching for new material.
- Same time every day: after breakfast, on the train (mumble quietly), or before bed.
- Missed a day? Do Day 7's review instead of starting over.
- Week 2: repeat the plan with a new short dialogue or video at the same level.
Common beginner mistakes and how to fix them
Most problems English speakers have with shadowing come from English habits. The good news: once you know the pattern, you can hear it in your own recording and fix it.
1. Using English stress rhythm
English speakers make one syllable loud and long: "ka-RA-te", "toy-OH-ta", "ha-JI-me-MA-shi-te". Japanese does not use loudness like this. Every mora is about the same length, and the melody comes from pitch (high and low), not from volume.
- Clap or tap once per mora: は・じ・め・ま・し・て = 6 taps, all even.
- Say the sentence like a quiet, steady drum, then add the melody from the audio.
2. Skipping long vowels and small っ
A long vowel is two beats, not one. A small っ is a short silent beat. If you skip them, you can say a different word: おばさん (obasan, aunt) vs おばあさん (obāsan, grandmother), or きて (kite, come) vs きって (kitte, stamp).
- Mark long vowels and っ in your script before you shadow (step 2).
- For っ, stop your mouth for one beat, as if you are about to say the next sound but wait.
- In your recording, count the beats in one word and compare with the model.
3. Reading romaji instead of kana
Romaji makes your brain use English spelling rules. "Kare" (he) becomes "care", "sake" becomes "sah-kee", and "Tokyo" loses its long vowels. Kana keeps you inside the Japanese sound system.
- Use kana or furigana scripts for shadowing. Use romaji only to check a reading you are not sure about.
- If you still need romaji, learn kana first. It usually takes a few weeks of daily practice.
4. Other common mistakes
| Mistake | Why it hurts | Fix |
|---|---|---|
| Material is too fast or too long | You fall behind and just mumble | Use slow audio and 3–4 sentences |
| Not knowing the meaning | You copy sounds, not language | Check every sentence in step 2 |
| Stopping after every mistake | It turns into repeating, not shadowing | Keep going; fix it in the next round |
| Never recording yourself | You cannot hear your own errors while speaking | Record once per session |
| Saying ん as a quick "n" glued to the next sound | Words lose a beat: こんな sounds like こな | Hold ん for a full beat |
Want more practice on these sounds? The Learn Japanese hub collects all the Japanese tools in one place: furigana subtitles, real-life and anime listening, AI pronunciation scoring and JLPT N5–N1 vocabulary.
FAQ
Can complete beginners do Japanese shadowing?
Yes, if you can read kana and the material is short and slow. Start with 3–4 sentences you understand, use slow audio, and shadow with the script first. If you cannot read kana yet, learn it first and use simple "listen and repeat" in the meantime.
How long should I shadow every day?
10–15 minutes a day is enough for beginners. Short daily sessions are better than one long session. Use the same passage for 2–3 days before moving on.
Should I shadow anime?
Anime can be fun later, but it is not ideal for beginners. Characters often use exaggerated voices, casual or rough speech, and fast speed. Start with natural, slow speech, then move to anime once you can shadow N4–N3 material comfortably.
Does shadowing improve listening too?
Yes. To shadow, you must hear every beat, including long vowels, っ and ん. Many learners find that words they practised in shadowing become easier to catch in real speech. It works best when you also know the meaning of what you shadow.
Is Corodomo free for shadowing practice?
Corodomo has a Free plan with a limited number of videos, up to 3 uploads and limited AI usage. Full AI features, including AI pronunciation scoring, are in the Plus plan (from $6/month billed yearly).
Conclusion
- Shadowing trains what textbooks cannot: mora rhythm, long vowels, っ, ん and natural intonation.
- Follow the 5 steps in order: listen, read along, shadow with the script, shadow blind, then record and compare.
- Keep it small and daily: 3–4 sentences, 15 minutes, slow audio first, and the same passage for 2–3 days.
Practice with Corodomo: Learn Japanese · Learn with videos · JLPT practice