Your mouth learns a language differently than your brain does. Shadow & Build trains the part that freezes.
Free Shadow & BuildYou already know more English than you can say. You can read it, you can understand a podcast, and then someone asks you a real question and the sentence falls apart on the way out of your mouth.
That is not a knowledge gap. It is a gap between what you can build in your head, slowly, with time to think, and what you can say out loud, at real speed, with someone waiting for your answer.
Grammar apps and vocabulary drills don't touch that gap. They train what you know. They don't train what you say.
Shadowing, listening to a native speaker and repeating right along with them, has a real research base behind why it works.
Researchers who study how adults learn languages (DeKeyser, Segalowitz) call this proceduralization: turning something you understand into something your mouth does without thinking first. That's the exact gap most of my students describe. They understand the grammar. Their mouth just hasn't practiced doing it yet.
Your phonological loop, the part of memory that holds sound just long enough to use it, gets stronger the more it repeats real, connected speech (Baddeley's working memory research, applied directly to shadowing by Kadota). That's why a sentence you've shadowed a dozen times comes out whole.
Native speech drops sounds, blends words together, and speeds up in ways a textbook sentence never does. Studies on shadowing (Tamai, Hamada) found it sharpens learners' ability to parse that, the fast, connected, real version of English, not the slow, textbook one.
Language researchers call the anxiety that blocks you from speaking the affective filter (Krashen). The less your mouth has to build from scratch under pressure, the less there is to freeze around. You're retrieving something you've already said out loud, successfully, many times before.
And it only works if what you're copying is real. Research on shadowing specifically points to gains in prosody, the rise and fall, stress, and rhythm of natural speech, not just the words themselves (Foote & McDonough, Mori). You can say every word correctly and still sound flat if you've never practiced the music underneath the sentence. That's why every recording in Shadow & Build is my own voice, not a computer reading text out loud. A synthesized voice has no real inflection to copy in the first place. You're shadowing an actual person, with actual rhythm, saying it the way it's really said.
"I can literally talk with a bit of confidence now, and I just keep getting stronger."
"I don't feel afraid to talk in public."
Every sentence is picked for a specific reason: it's something you're actually going to need, in a situation you're actually going to be in. A phone call you'll have to make. A question about your weekend you'll have to answer without stalling. Not "the weather is nice today."
Every sentence is recorded in my own voice, with my own inflection, not a synthesized one. You start with the last few words and repeat them until they're smooth, then add a few more words before that, working backward until you're saying the whole sentence in one breath, with the same stress and rhythm I used. Language teachers call this back-chaining, and it's especially good at fixing the parts of a sentence that get dropped or blurred together when someone actually talks fast.
After enough repetition, you're not constructing the sentence anymore. You're just saying it, in the same rhythm you practiced. That's the whole goal.
A free group for women learning English, where the next pack unlocks and you're not doing this alone.
Join the CommunityIf you want to go deeper, I teach 1:1 lessons built around your actual life, not a textbook.
Book a Trial LessonDeKeyser, R. M. (2007). Skill acquisition theory. In VanPatten & Williams (Eds.), Theories in Second Language Acquisition. Proceduralization, turning known grammar into automatic speech.
Segalowitz, N. (2010). Cognitive Bases of Second Language Fluency. Routledge. Fluency as an automatized skill, not just knowledge.
Baddeley, A. (2003). Working memory and language: an overview. Journal of Communication Disorders. The phonological loop and how it holds sound in short-term memory.
Kadota, S. (2012). Shadowing as a Practice in Second Language Acquisition. Routledge. Applies working-memory research directly to shadowing practice.
Tamai, K. (1992) and Hamada, Y. (2016). Shadowing and its effect on parsing fast, connected native speech.
Foote, J. A., & McDonough, K. (2017). Using shadowing with mobile technology to improve L2 pronunciation. Journal of Second Language Pronunciation. Prosody and pronunciation gains from shadowing.
Mori, Y. (2011). Shadowing with oral reading: Effects of combined training on the improvement of Japanese EFL learners' prosody. Language Education & Technology.
Krashen, S. (1982). Principles and Practice in Second Language Acquisition. The affective filter, anxiety's role in blocking speech.
Start with the app, free, in my own voice. See where it takes you.