Shadowing (repeating a recording a fraction of a second behind the speaker) trains real-time speech processing, prosody and fluency at once, needs almost no preparation, and works with any short authentic audio.
Procedure
- 30–60 seconds of audio; listen for gist.
- Transcript; clarify vocabulary; shadow while reading.
- Repeat focusing on rhythm and stress, not words; remove the transcript.
- Shadowing ladder: echo → supported (transcript) → guided (no transcript) → natural (authentic audio) → performance (record, compare, repeat over days).
- Week routine: Mon listen for meaning · Tue shadow with transcript · Wed without · Thu record and compare · Fri repeat.
- Avoid: fast recordings, too much unknown vocabulary, correcting mid-shadow.
Sources listed
Kadota (2019) Shadowing as a practice in SLA; Tamai (2005); Hamada (2018); Field (2008); Gilbert (2008); Underhill (2005); Nation & Newton (2009).
Research note (Claude)
Most shadowing studies are Japanese EFL university samples (Hamada; Kadota), showing gains in bottom-up decoding and listening scores more clearly than in speaking. Treat it primarily as a listening/phonological-loop exercise with a pronunciation bonus.
Apply in my context (Claude)
- IELTS Speaking: shadow a band-8 Part 2 sample for prosody, then deliver your own answer on the same topic.
Connections
Learners decode far less than teachers think (fuzzy lexical representations) · Bottom-up listening instruction alone is not enough · Visualise sentence stress with audiograms