LearnAnyLanguage.com

How do I get more speaking practice when learning a language?

Evidence-based answer · Last updated · How it’s made

The short answer

The research is general second-language-acquisition evidence (none of it specific to Norwegian), but it consistently points the same way: speaking skill grows through actual interaction — real or simulated conversation, tasks that require you to negotiate meaning, and repeated attempts with feedback — more than through passive study. You can get more of this practice through classroom-style tasks, conversation partners, drama/role-play, or newer tools like chat apps, voice assistants, and AI chat partners, but your own willingness to jump into a conversation despite discomfort is itself a major factor in how much benefit you get.

Asked as someone learning Norwegian (beta). Findings drawn from Norwegian-specific research are framed for Norwegian learners and may not hold for other languages; where the evidence is general second-language research, the report says so.

Norwegian has a thinner research base than our core languages, so this answer may lean on general second-language findings.

Interaction drives speaking gains, robustly

Decades of research and multiple meta-analyses show that interacting in the language you're learning (having conversations, negotiating misunderstandings, getting feedback) produces measurably better language outcomes than not interacting, with the benefit showing up even more clearly weeks later than right after the practice.1

Task-based speaking practice beats conventional drills

Structuring speaking practice around meaningful tasks (solving a problem together, completing a real communicative goal) rather than grammar drills produced significantly bigger fluency gains in classroom experiments, both in-person and via mobile apps.4,5

In plain terms

'Task-based' means giving learners a real goal to accomplish using the language — like planning a trip together — instead of just practicing sentence patterns.

Example

For example, one study had students go through six sessions of task-based speaking activities and their fluency score jumped from 2.63 to 3.87 out of 5, far more than a comparison group taught conventionally.

Willingness to communicate matters as much as opportunity

How much a learner actually speaks up — not just how many chances they're given — strongly predicts how much their spoken language improves. Learners who are more willing to jump into conversation tend to produce more speech and improve faster, and this willingness can be increased by things like choosing topics learners find personally relevant, building confidence, and creating a low-pressure atmosphere.2,12,13,14,16

In plain terms

Researchers call this trait 'willingness to communicate' (WTC) — basically how ready someone feels to actually speak up in the moment, which is separate from their raw ability to speak.

Drama and role-play build confidence to speak

Structured drama-based activities and role-play — preparing a scenario, acting it out, then reflecting on it — have been shown to increase learners' oral proficiency and their willingness to communicate, partly by lowering anxiety and making practice feel like play rather than a test.2,6

Example

For instance, one program had students research a scenario, write a script together, rehearse, perform it, and then discuss how it went — over just three weeks this improved learners' handling of conversation flow and task completion.

Mobile chat apps and technology-mediated conversation help, with caveats

Practicing speaking through computer- or phone-mediated communication (text chat, voice chat, WhatsApp-style group tasks) produces a moderate positive effect on oral proficiency compared to no interaction at all, but a meta-analysis found it can sometimes hurt fluency and accuracy compared to face-to-face conversation, and results depend heavily on the type of task used.3,8,15

AI conversation partners can help, but novelty fades

AI chatbots, smart speakers, and AI-assisted role-play tools can give learners extra low-pressure speaking practice and modest proficiency gains, especially useful when human partners aren't available. However, learners' enthusiasm for AI partners tends to cool over time because AI lacks natural back-and-forth, tone, and genuinely responsive feedback, and learners end up preferring real people for authentic conversation while using AI more narrowly for things like building fluency or vocabulary.7,10,11,17

Virtual reality can lower anxiety and simulate rehearsal

Immersive virtual reality speaking tasks can make learners feel more present and engaged and can serve as a low-stakes 'rehearsal space' for real conversations, but in some virtual/metaverse settings the pressure of being observed and evaluated in real time can actually increase speaking anxiety and hurt performance.9,18

What to do with this

  • None of this evidence is specific to Norwegian, so the guidance below is general and should transfer to Norwegian learning rather than being Norwegian-specific.
  • First, prioritize actual back-and-forth interaction over solo drilling: find a conversation partner, tutor, or language exchange partner and use tasks with a real goal (planning something, solving a problem, debating a topic) rather than open-ended chit-chat, since goal-based tasks produced the largest fluency gains in the studies reviewed.
  • Second, don't wait to feel "ready" — the evidence suggests that simply pushing yourself to speak up more, even imperfectly, is one of the biggest levers on improvement, so pick topics you personally care about and lower the stakes (small groups, familiar partners, drama/role-play framing) to make that easier.
  • Third, treat text/voice chat apps, AI chatbots, and voice assistants as a useful supplement for extra reps when a human partner isn't available, but not a full replacement for face-to-face conversation, since technology-mediated speaking can dip in fluency/accuracy compared to talking with a real person and AI partners tend to feel less satisfying over time.
  • Finally, if anxiety is holding you back, low-pressure formats like rehearsed role-play or VR practice can help you build confidence before higher-stakes real conversations, but be aware that some tech-mediated formats with real-time observation can increase anxiety rather than reduce it, so it is worth noticing which format actually calms you versus which one raises your stress.

Now pick something to do it with

Each of these opens an overview of the apps, courses and immersion material worth a look for that — filtered to exactly what the advice above points you at:

Conversation practiceLanguage exchangeGroup lessonsPrivate tutoring
Or see the full resource list

Worth knowing

  • All sources here are general second-language-acquisition studies — of English, Spanish, French, and Chinese learners, mostly in classroom or lab settings — and none study Norwegian specifically.
  • There is no reason from these sources to think Norwegian speaking practice works fundamentally differently, but nothing here confirms it either, so treat this as general SLA guidance rather than Norwegian-tailored advice.
  • Many of the studies are small (dozens of students), short-term, or based on quasi-experimental classroom designs rather than large randomized trials, and several results (on computer-mediated communication, AI chatbots) are described by the researchers themselves as exploratory or limited by sample size.
  • Findings on willingness to communicate are mostly correlational, showing that more willing speakers tend to improve more, without fully proving that increasing willingness alone causes the improvement.

Want a deeper literature dive?

There’s enough published research here to go wider than the usual pass. Available on the Pro plan.

See plans

  1. [1]S. Loewen, Masatoshi Sato. Interaction and instructed second language acquisition. Language Teaching 2018. doi.org/10.1017/S0261444818000125
  2. [2]Kent Lee, Marilyn L. Abbott, N. Chen. Increasing Students’ Willingness to Communicate: Drama-Based Approaches to Language Instruction in English for Academic Purposes Classes. The TESL Canada Journal 2020. doi.org/10.18806/TESL.V37I3.1346
  3. [3]Sherine Akkara, Venkata Surya Seshagiri Anumula, Mallikarjuna Sastry Mallampalli. Impact of WhatsApp Interaction on Improving L2 Speaking Skills. International Journal of Emerging Technologies in Learning (iJET) 2020. doi.org/10.3991/ijet.v15i03.11534
  4. [4]Khofifah Indah Baitul Jannah, Siti Ulfya Syahara Razaq, Hafizudin Hafizudin. Revitalizing Speaking Fluency through Task-Based Language Teaching: Addressing Post-Pandemic Communicative Gaps in EFL Classrooms. Bulletin of Scientific Reseach in English Education 2025. doi.org/10.59261/bsree.v2i1.8
  5. [5]Shijiao Jia, Zhaoxia Lu. Enhancing EFL Oral Production through Mobile-Assisted Task-Based Language Teaching: A Study in Effectiveness. J. Inf. Technol. Educ. Res. 2025. doi.org/10.28945/5482
  6. [6]C. Man, Andrea Lee Jen May, Sabariah Sharif, S. Singh. Exploring the Use of Drama-Based Activities as a Learning Tool in English Oral Proficiency Development among Students from Non-Malay-Medium Nation Schools: A Preliminary Study. International Journal of Academic Research in Business and Social Sciences 2019. doi.org/10.6007/ijarbss/v9-i6/6082
  7. [7]Gavin Bui. From Novelty to Strategy: The Trajectory of Learner Perceptions in Human and AI-Assisted L2 Speaking Tasks. International Journal of TESOL Studies 2026. doi.org/10.58304/ijts.260301
  8. [8]Huifen Lin. Computer-mediated communication (CMC) in L2 oral proficiency development: A meta-analysis. ReCALL 2014. doi.org/10.1017/s095834401400041x
  9. [9]Zafer Üstünbaş, Esim Gursoy. Immersive Virtual Reality as a Rehearsal Space for L2 Speaking: EFL Learners’ Perceived Social Presence, Situational Motivation, and Willingness to Communicate. Journal of language research 2026. doi.org/10.51726/jlr.1906805
  10. [10]Darmawansah Darmawansah, Gwo-Jen Hwang, Chi-Jen Lin, Febiyani Febiyani. An artificial intelligence-supported GFCA learning model to enhance L2 students’ role-play performance, English speaking and interaction mindset. Educational technology research and development 2025. doi.org/10.1007/s11423-025-10453-6
  11. [11]Gilbert Dizon. Evaluating intelligent personal assistants for L2 listening and speaking development. Language learning & technology 2020. doi.org/10.64152/10125/44705
  12. [12]Mackenzie Coulter-Kern. Examining Willingness to Communicate and Complexity, Accuracy, Fluency, and Functional Adequacy within Task-Based Interaction. ProQuest LLC 2023. http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&res_dat=xri:pqm&rft_dat=xri:pqdiss:30570367
  13. [13]Graham Robson. The Relationship between WTC and Oral Proficiency Measurements in the Study Abroad Context.. 2015. doi.org/10.5539/IES.V8N12P56
  14. [14]Robson, Graham G.. The Relationship between WTC and Oral Proficiency Measurements in the Study Abroad Context. International Education Studies 2015. https://eric.ed.gov/?id=EJ1083905
  15. [15]Scott Payne, B. M. Ross. Synchronous CMC, Working Memory, and L2 Oral Proficiency Development. Language Learning & Technology 2005. doi.org/10.64152/10125/44031
  16. [16]Paul Leeming, Joseph P. Vitta, Phil Hiver, Dillon Hicks, Stuart Mclean, Christopher Nicklin. Willingness to Communicate, Speaking Self‐Efficacy, and Perceived Communicative Competence as Predictors of Second Language Spoken Task Production. Language Learning 2024. doi.org/10.1111/lang.12640
  17. [17]E. Verhelst, Tony Belpaeme. Large Language Models Cover for Speech Recognition Mistakes: Evaluating Conversational AI for Second Language Learners. IEEE/ACM International Conference on Human-Robot Interaction 2025. doi.org/10.1109/HRI61500.2025.10974188
  18. [18]Ferdi Çelik, Meltem Huri Baturay, Ahmet Erdost Yastıbaş, Ehsan Namaziandost, C. Ersanlı. Modelling the complex dynamic effect of L2 speaking anxiety in metaverse on students' academic enjoyment, pronunciation anxiety, speaking performance, and task load. British Educational Research Journal 2025. doi.org/10.1002/berj.70062