Why Your English Sound Rules Follow You (and How to Retrain Them)

Much of an English speaker's accent in Spanish, French, Korean or Portuguese comes from sound rules English runs automatically: a puff of air after p, t, k, schwa in unstressed syllables, a dark L and the English R. Those rules keep running in the new language. To retrain them, build the new sound category by ear, then suppress the old rule in your mouth.

First, a ground rule. ASHA puts it plainly: "Accents are NOT a communication disorder" (ASHA). The goal here is intelligibility and confidence, not accent elimination.

Sound rules, not just sounds

Speech-language pathologists separate two sides of speech sounds, as ASHA's Practice Portal describes: the motor side (articulation) and the linguistic side, meaning the patterns and rules of a sound system (phonology). Our motor learning guide covers the first. This guide covers the second. ASHA's guidance, written about children who speak more than one language or dialect, notes that the rules of one system may transfer to another, and that these influences do not indicate a speech sound disorder.

You can watch an English rule work inside English. The p in pin has a puff of air. The p in spin doesn't. The same goes for key and ski (Broeders & Gussenhoven). You never learned that rule on purpose. You just run it. That's why it follows you.

Four English rules that follow you

Aspiration

English voiceless stops at the start of a word are aspirated, with roughly 30 to 120 milliseconds before voicing starts. Spanish keeps that gap between 0 and 30 milliseconds (Amengual, 2023). This rule is sticky. In the same study, British-born English speakers who had lived in Spain for more than 30 years still produced Spanish p, t, k with compromise values between English and Spanish. French voiceless stops are unaspirated too (Fougeron & Smith, 1993), and so are Brazilian Portuguese ones (Kupske & de Oliveira, 2020). Korean is the twist: aspiration is meaningful there, alongside pitch.

Vowel reduction

English shrinks unstressed vowels. The first o in photograph is a full vowel. In photography, it reduces to schwa. Researchers call vowel reduction a prominent feature of American English (Byers & Yavas, 2017). Spanish has five vowels, and each one occurs in both stressed and unstressed syllables (Martínez-Celdrán, Fernández-Planas & Carrera-Sabaté, 2003). In a study of 60 adult learners of Spanish, all learners produced unstressed vowels with some centralization, a sign of the English rule at work. Advanced learners' vowels were close to native (Menke & Face, 2010). Portuguese is different again: in Brazilian Portuguese, final unstressed o sounds close to [u], as in mato [ˈma.tʊ]. The target isn't "never reduce." It's "reduce the way the new language does."

Dark L

Many English accents darken L at the end of a syllable, as in feel or milk. Spanish uses only a subset of English's L variants, and learners' Spanish L moves toward native norms with study (Solon, 2017). Brazilian Portuguese goes another way: in most Brazilian accents, L at the end of a syllable becomes [w], so Brasil is [bɾa.ˈziw] (Barbosa & Albano, 2004). Korean ㄹ at the end of a word is a lateral [l], as in 달 [tal] "moon."

English R

English R is usually an approximant [ɹ]: the tongue gets close to the roof of the mouth without tapping it. Spanish uses a tap and a trill. French most often uses [ʁ], made far back in the mouth, as in rue [ʁy] (Fougeron & Smith, 1993). Brazilian Portuguese uses a tap in caro and, for the "strong" r in carro, usually a sound like [x] or [h] (Barbosa & Albano, 2004). Korean ㄹ is a flap [ɾ] between vowels and [l] at the end of a word.

Bonus: American English also has a rule that helps you. The t or d between vowels in butter or ladder becomes a flap, and that flap is very close to the Spanish tap (Daidone & Darcy, 2014). Keep that rule. Use it on purpose.

Why similar sounds are the hardest: two models

Two research models explain why some new sounds come easily and others never seem to settle.

Sources: Flege & Bohn, 2021; Best & Tyler, 2007; Tyler, 2021.

One of the SLM's key claims is that a similar second-language sound can merge with its first-language neighbor, while a new sound has no such link and is often produced more accurately (Sypiańska & Constantin, 2021). Classic evidence comes from French: English-speaking learners produced the new vowel [y] more accurately than the familiar-seeming [u] (Flege, 1987; summarized by Oakley, 2019).

Two more findings worth knowing. The influence runs both ways: adult English speakers in elementary Korean classes showed shifts in their English speech toward Korean patterns after brief Korean experience (Chang, 2012). And hearing a contrast isn't the same as storing it in your words. English-speaking learners of Spanish could discriminate the tap and trill, yet still accepted nonwords with the wrong one as real words (Daidone & Darcy, 2014).

Step 1: Train your ears

SLPs use speech sound perception training to help a child build a stable perceptual representation of a target sound (ASHA). Second language research has a well-tested adult version: high-variability phonetic training (HVPT). You hear a contrast in many words, from many speakers, decide which sound you heard, and usually get immediate feedback.

The method grew out of a 1991 study in which training Japanese listeners with many natural recordings of English r and l worked better than earlier methods (Logan, Lively & Pisoni, 1991). In a later study, that perceptual learning carried over to the learners' own pronunciation (Bradlow et al., 1997). A 2025 meta-analysis of 79 studies found medium-to-large gains in perception that lasted, with some transfer to new items; the number of talkers and total training time mattered (Uchihara, Karas & Thomson, 2025). Another meta-analysis found that perception training produced small but real gains in production too (Sakai & Moorman, 2018).

One caution: early on, keep listening blocks and speaking blocks separate. In lab studies, producing the sounds during perception training sometimes disrupted perceptual learning (Baese-Berk & Samuel, 2016).

Step 2: Break the rule with contrast drills

SLPs teach new sound categories through contrast. ASHA describes several contrast approaches, including these three (ASHA Practice Portal). They were developed for children's speech, so treat the adult versions below as borrowed logic, not tested treatments.

In Gierut's case report, treating only three sets of maximal contrasts led a child to learn 16 word-initial consonants. Contrast can do more than fix one word at a time.

Step 3: Rotate targets with cycles

Hodson's cycles approach targets error patterns in rotation. ASHA describes treatment scheduled in cycles of 5 to 16 weeks, with one or more patterns targeted in each. Patterns are recycled until they emerge in spontaneous speech, and the aim is to stimulate emergence, not to force mastery before moving on (ASHA).

For an adult learner, that becomes a simple plan. Spend one week on one rule, such as aspiration. Next week, vowels. Then L, then R. Then start the cycle again. Rotating fits the spacing principle from motor learning.

Step 4: Put it back in connected speech

Drills are not the finish line. Gains can carry over, though. Learners of Spanish who trained voice onset time with visual feedback, using only short carrier sentences, improved in more continuous and spontaneous speech too (Offerman & Olson, 2016). Retelling a story is a natural transfer test. Your rules either hold up in real sentences, or they don't yet.

FAQ

Is a foreign accent a speech disorder? No. ASHA states that accents are not a communication disorder, and it describes accent modification as elective. The useful goals are being understood and feeling confident.

Will I ever lose my accent? You don't need to. A strong accent does not necessarily make speech harder to understand (Munro & Derwing, 1995). Focus on the rules that change meaning, like pero vs. perro, and on the ones listeners notice most.

Why can I hear the difference but not say it? Perception and production are linked, but loosely. Perception training improves production a little (Sakai & Moorman, 2018), and gains in the two don't always move together (Kartushina et al., 2015). Learners can also hear a contrast without storing it correctly in the words they know.

What is high-variability phonetic training? HVPT is ear training with many words and many speakers. You hear a contrast, choose which sound you heard and usually get immediate feedback. Meta-analyses show lasting gains in perception and smaller gains in pronunciation.

Which sounds should I work on first? Start with contrasts that change meaning: the Spanish tap and trill, French [y] and [u], Korean lax, tense and aspirated stops, and Portuguese oral and nasal vowels. Then work on rules that affect many words at once, like aspiration and vowel reduction.

Can learning a new language change my English? A little, yes. Adult English speakers in beginning Korean classes showed measurable shifts in their English speech after brief Korean study (Chang, 2012).

Your next step: Rules show up in real speech. Try a story retell in Edgewise and put your new habits to work. Edgewise scores story structure and sentence building, sets your level from your first retell and ends each round with your next piece: one specific thing to add next time.

Join early access

Related: Speech therapy techniques for language learners · How to practice pronunciation with motor learning

Sources

1. Amengual, M. (2023). The acoustic realization of L2 Spanish phonetic categories and allophonic alternations by English-speaking immigrants. Proceedings of the 20th International Congress of Phonetic Sciences (ICPhS 2023, Prague), paper 154. https://www.internationalphoneticassociation.org/icphs-proceedings/ICPhS2023/full_papers/154.pdf
2. American Speech-Language-Hearing Association. (n.d.). Accent modification [Practice Portal]. https://www.asha.org/practice-portal/professional-issues/accent-modification/
3. American Speech-Language-Hearing Association. (n.d.). Speech sound disorders: Articulation and phonology [Practice Portal]. https://www.asha.org/practice-portal/clinical-topics/articulation-and-phonology/
4. Baese-Berk, M. M., & Samuel, A. G. (2016). Listeners beware: Speech production may be bad for learning speech sounds. Journal of Memory and Language, 89, 23–36. https://doi.org/10.1016/j.jml.2015.10.008
5. Barbosa, P. A., & Albano, E. C. (2004). Brazilian Portuguese. Journal of the International Phonetic Association, 34(2), 227–232. https://doi.org/10.1017/S0025100304001756
6. Best, C. T., & Tyler, M. D. (2007). Nonnative and second-language speech perception: Commonalities and complementarities. In O.-S. Bohn & M. J. Munro (Eds.), Language experience in second language speech learning: In honor of James Emil Flege (pp. 13–34). John Benjamins. https://doi.org/10.1075/lllt.17.07bes
7. Bradlow, A. R., Pisoni, D. B., Akahane-Yamada, R., & Tohkura, Y. (1997). Training Japanese listeners to identify English /r/ and /l/: IV. Some effects of perceptual learning on speech production. The Journal of the Acoustical Society of America, 101(4), 2299–2310. https://doi.org/10.1121/1.418276
8. Broeders, T., & Gussenhoven, C. (n.d.). 9.2 Aspiration. In An introduction to American English phonetics. University of Groningen Open Textbooks. https://opentextbooks.rug.nl/americanenglishphonetics2/chapter/9-2-aspiration/
9. Byers, E., & Yavas, M. (2017). Vowel reduction in word-final position by early and late Spanish-English bilinguals. PLOS ONE, 12(4), e0175226. https://doi.org/10.1371/journal.pone.0175226
10. Chang, C. B. (2012). Rapid and multifaceted effects of second-language learning on first-language speech production. Journal of Phonetics, 40(2), 249–268. https://doi.org/10.1016/j.wocn.2011.10.007
11. Daidone, D., & Darcy, I. (2014). Quierro comprar una guitara: Lexical encoding of the tap and trill by L2 learners of Spanish. In R. T. Miller et al. (Eds.), Selected proceedings of the 2012 Second Language Research Forum (pp. 39–50). Cascadilla Proceedings Project. https://www.lingref.com/cpp/slrf/2012/paper3084.pdf
12. Flege, J. E. (1987). The production of "new" and "similar" phones in a foreign language: Evidence for the effect of equivalence classification. Journal of Phonetics, 15(1), 47–65. https://doi.org/10.1016/S0095-4470%2819%2930537-6
13. Flege, J. E. (1995). Second language speech learning: Theory, findings, and problems. In W. Strange (Ed.), Speech perception and linguistic experience: Issues in cross-language research. York Press. No open URL; summarized in Flege & Bohn (2021).
14. Flege, J. E., & Bohn, O.-S. (2021). The revised speech learning model (SLM-r). In R. Wayland (Ed.), Second language speech learning: Theoretical and empirical progress (pp. 3–83). Cambridge University Press. https://doi.org/10.1017/9781108886901.002
15. Fougeron, C., & Smith, C. L. (1993). French. Journal of the International Phonetic Association, 23(2), 73–76. https://doi.org/10.1017/S0025100300004874
16. Gierut, J. A. (1989). Maximal opposition approach to phonological treatment. Journal of Speech and Hearing Disorders, 54(1), 9–19. https://doi.org/10.1044/jshd.5401.09
17. Kartushina, N., Hervais-Adelman, A., Frauenfelder, U. H., & Golestani, N. (2015). The effect of phonetic production training with visual feedback on the perception and production of foreign speech sounds. The Journal of the Acoustical Society of America, 138(2), 817–832. https://doi.org/10.1121/1.4926561
18. Kupske, F. F., & De Oliveira, M. S. (2020). O desenvolvimento do padrão de voice onset time das oclusivas surdas iniciais do inglês por aprendizes soteropolitanos: Efeitos da instrução explícita [In Portuguese]. Ilha do Desterro, 73(3), 185–204. https://doi.org/10.5007/2175-8026.2020v73n3p185
19. Logan, J. S., Lively, S. E., & Pisoni, D. B. (1991). Training Japanese listeners to identify English /r/ and /l/: A first report. The Journal of the Acoustical Society of America, 89(2), 874–886. https://doi.org/10.1121/1.1894649
20. Martínez-Celdrán, E., Fernández-Planas, A. M., & Carrera-Sabaté, J. (2003). Castilian Spanish. Journal of the International Phonetic Association, 33(2), 255–259. https://doi.org/10.1017/S0025100303001373
21. Menke, M. R., & Face, T. L. (2010). Second language Spanish vowel production: An acoustic analysis. Studies in Hispanic and Lusophone Linguistics, 3(1), 181–214. https://doi.org/10.1515/shll-2010-1069
22. Munro, M. J., & Derwing, T. M. (1995). Foreign accent, comprehensibility, and intelligibility in the speech of second language learners. Language Learning, 45(1), 73–97. https://doi.org/10.1111/j.1467-1770.1995.tb00963.x
23. Oakley, M. (2019). Articulation of L2 French mid and high vowels. Proceedings of the 19th International Congress of Phonetic Sciences (ICPhS 2019, Melbourne). https://assta.org/proceedings/ICPhS2019Microsite/pdf/full-paper_414.pdf
24. Offerman, H. M., & Olson, D. J. (2016). Visual feedback and second language segmental production: The generalizability of pronunciation gains. System, 59, 45–60. https://doi.org/10.1016/j.system.2016.03.003
25. Sakai, M., & Moorman, C. (2018). Can perception training improve the production of second language phonemes? A meta-analytic review of 25 years of perception training research. Applied Psycholinguistics, 39(1), 187–224. https://doi.org/10.1017/S0142716417000418
26. Solon, M. (2017). Do learners lighten up? Phonetic and allophonic acquisition of Spanish /l/ by English-speaking learners. Studies in Second Language Acquisition, 39(4), 801–832. https://doi.org/10.1017/S0272263116000279
27. Sypiańska, J., & Constantin, E.-R. (2021). New vs. similar sound production accuracy: The uneven fight. Yearbook of the Poznań Linguistic Meeting, 7(1), 155–179. https://doi.org/10.14746/yplm.2021.7.7
28. Tyler, M. D. (2021). Perceived phonological overlap in second-language categories: The acquisition of English /r/ and /l/ by Japanese native listeners. Languages, 6(1), 4. https://doi.org/10.3390/languages6010004
29. Uchihara, T., Karas, M., & Thomson, R. I. (2025). High variability phonetic training (HVPT): A meta-analysis of L2 perceptual training studies. Studies in Second Language Acquisition, 47(3), 794–827. https://doi.org/10.1017/S0272263125100879
30. Weiner, F. F. (1981). Treatment of phonological disability using the method of meaningful minimal contrast: Two case studies. Journal of Speech and Hearing Disorders, 46(1), 97–103. https://doi.org/10.1044/jshd.4601.97
31. Williams, A. L. (2000). Multiple oppositions: Theoretical foundations for an alternative contrastive intervention approach. American Journal of Speech-Language Pathology, 9(4), 282–288. https://doi.org/10.1044/1058-0360.0904.282

IPA note: example transcriptions were checked against Wiktionary entries and Wikipedia phonology articles that cite the Journal of the International Phonetic Association Illustrations for each language.