English shadowing, explained.

You play a short line of English, pause, and say it back the way you heard it: the same speed, the same stress, the same pauses. You are not studying the sentence; you are copying a person. Interpreters train this way, and it works for learners for the same reason — your mouth learns against a model instead of against a rule.

The gap between what you heard and what you said is the whole lesson. Most learners never hear that gap, because they have nobody to compare against. Shadowing puts the model and your attempt side by side, line after line.

Listen. Play one line — one, not the whole clip — and listen for where the voice rises, which word is stressed, and where the speaker breathes.

Your turn. Say it back at once, out loud, at the speaker's pace. Do not read it first. If it is too fast, slow the clip down and say it slowly; then bring the speed back up.

Compare. Play the line again and listen for the difference. A sound you flattened, a pause you skipped, a word you put weight on that the speaker did not.

Repeat, then move on. Two or three tries, then the next line. Five minutes a day of this beats an hour on Sunday.

Real speech, not textbook audio. Textbook recordings are slow, even and over-articulated, so shadowing them teaches you an English nobody speaks. Short-form videos are ideal to start with: thirty seconds, one speaker, and the sentence-length lines shadowing needs.

Podcasts are the next step. They are longer and looser, and with a full transcript you can listen along, then pick a passage and shadow it line by line. Choose a speaker you would like to sound like, and stay with them for a while; an accent is easier to copy from one voice than from twenty.

Rhythm first. English stresses a few words in a sentence and hurries over the rest, and that pattern is most of what makes speech sound natural or not. Then linking: native speakers run words together — “an apple” is one movement, not two words — and pause where the meaning pauses.

Last, individual sounds. Every learner has a few of the 41 sounds of American English that their mother tongue does not have, and those few are what people hear as an accent. Shadowing shows you which ones, if something is measuring them.

The app is built around this loop. A Short plays one line, stops and waits for your turn, and scores what you said — every word coloured green, orange or red, with the sound inside a weak word named. Any line can be slowed to 0.8x or 0.6x, and every line has a translation under it in your own language.

Podcasts come with the complete transcript and play as video or audio only, in the background while you walk. Turn on My voice and the clip plays back with your own takes where the original lines were, so you hear yourself where the native speaker was. And every line you shadow updates your pronunciation score across all 41 sounds.