Maintenance: the shadowing protocol

Unité PO-34 · campagne 22 · borrow a native mouth for ten minutes.

Shadowing is imitation, assumed and total: play a short native extract and speak right behind it, half a second late, copying everything — rhythm, liaisons, melody, even the breaths. You are not analysing; you are letting your mouth copy what your ear hears. Ten focused minutes on one 30-second extract beats an unfocused hour.

The protocol:

1. Pick a 30-second extract you like (podcast, interview — anything with one clear voice).
2. Three passes listening, eyes closed, no speaking.
3. Five passes shadowing — speak over/behind the voice, matching its débit. Falling behind? Drop words, keep the melody: rhythm first, completeness later.
4. One pass alone: say the passage without the model.

Then verify the transfer here: say the two model sentences below (both native-voiced — listen, shadow, then record) and watch the liaison and rhythm scores — they are what shadowing trains.

Checking your progress...