CELBAN Speaking: 35 Minutes of Role Play Over Zoom

Most speaking exams give you between ten and twenty minutes. Michigan's MET allocates ten. Cambridge B2 First runs fourteen for a pair.

Canada's nursing English exam gives one candidate up to 35 minutes, conducted over Zoom, with role play prompts on screen.

SpeakShark is free to build the one habit that test is made of, three AI conversation sessions a day with no card.

In this guide: what CELBAN is · the length · role play · double rating · the five minute rule · preparing · limits · method · FAQ

Key takeaways

• CELBAN is a nursing specific-purpose English exam used in Canada.
• The speaking test runs up to 35 minutes, longer than any exam we have catalogued.
• It is conducted over Zoom with a wired headset required.
• It mixes questions with role play, and the prompt stays on screen.
• It is recorded so a second examiner can rate it independently.

A nursing specific-purpose language exam

The CELBAN Centre describes the exam as assessing language tasks in the context of nursing, and puts it plainly: preparing for CELBAN is preparing for nursing in real life.

It calls CELBAN a unique nursing specific-purpose language exam, informed by nursing and language testing professionals, reflecting authentic nursing communication. It evaluates listening, writing, reading and speaking.

That framing matters because it is rare. Most English exams are general instruments that a profession then borrows. IELTS was not built for nurses. The Michigan English Test is accepted by more than thirty five American state nursing boards and is a general test. OET is the other profession-built exam, and CELBAN sits in the same small family.

Thirty five minutes is an outlier

Set the CELBAN speaking test against the speaking sections we have catalogued from published sources.

Every other long speaking test on that list gets its length by putting more people in the room. ECPE reaches thirty minutes with three candidates sharing it, which we set out in ECCE against ECPE.

CELBAN reaches thirty five with one candidate. That is not a longer exam so much as a fundamentally larger sample of one person's speech, which is what you would design if the decision being made was whether somebody can be trusted to communicate on a ward.

Role play with the prompt left on screen

The interview is not a question and answer session. The CELBAN Centre states that during the interview the examiner will have examinees answer questions and engage in role play.

The delivery detail is the useful part. The examiner shares role play prompts on the screen so examinees can review the information required to prepare their responses to selected tasks, and those prompts remain visible on the screen for the duration of the task.

Two things follow from that.

You are not being tested on memory. The information you need stays in front of you, which removes the failure mode where a candidate loses the thread of a scenario and cannot recover.

And you are being tested on using written information in speech, live, with a person responding. That is a different skill from describing a picture or giving an opinion, and it is much closer to what a nurse actually does on a shift: read a note, then tell somebody what it means.

Recorded so a second examiner can rate it

The CELBAN Centre states that the interview is recorded so that a second trained examiner can rate it independently, as well as to verify the score if needed and for future quality assurance checks.

Independent double rating is a serious reliability commitment. It is also the same principle Cambridge applies with two examiners in the room, though it splits their jobs rather than duplicating them: the interlocutor marks one criterion and the silent assessor marks the rest, which we covered in the two examiner roles.

For a candidate, the practical reading is that a single examiner having a bad day is not the whole story, and that the recording exists.

The five minute rule

One operational rule deserves its own line because it ends applications rather than costing marks.

Examinees are asked to arrive on time to allow the examiner to complete check-in procedures. If examinees arrive later than five minutes after their scheduled appointment time, they will not be allowed to complete the test.

Five minutes. Not fifteen, not a grace period at the examiner's discretion as far as the published wording goes.

The technical requirement sits alongside it. A wired headset must be plugged into the computer before the examinee enters the appointment, so that the examinee can clearly hear the examiner and the examiner can clearly hear the examinee. Discovering a headset problem at the appointment time is a five minute problem.

What this changes about preparing

• Build stamina for a long speaking turn. Thirty five minutes alone is a different endurance problem from fourteen minutes shared with a partner.
• Practise speaking from a written prompt rather than from memory, since the prompt stays visible.
• Practise role play specifically, because it is named as part of the interview.
• Test the headset the day before, not on the day.
• Treat the appointment time as hard. Five minutes late is over.

Role play against a visible prompt with somebody responding is the practice that matches this exam, which is why role play scenarios and a daily speaking partner are more useful here than a vocabulary list. Speaking practice for nurses covers the clinical vocabulary side, and testing your speaking level free gives you a starting point.

For the other routes internationally educated nurses take, MET for US nurse licensure covers a system that varies state by state, and Irish nursing English covers a regulator that lets you choose which skill drops. OET speaking criteria covers the other profession-built exam. For how examiners mark spoken English in general, Cambridge speaking criteria by level and Interactive Communication use real scored candidates, and CEFR levels for speaking explains the scale.

What we could not verify

We did not find the required benchmark score. Canadian nursing regulators are provincial, and the level a candidate must reach is set by the college they apply to rather than by the CELBAN Centre. We have not read a provincial requirement.

We did not read the other three sections. Listening, reading and writing have their own pages that we have not opened, so nothing here describes them.

We did not verify fees, availability or how often it runs.

The 35 minutes is a maximum. The CELBAN Centre says the test takes up to 35 minutes, not that it always runs that long.

We did not verify which provinces accept CELBAN, only that the CELBAN Centre describes it as the language test of choice for many internationally educated nurses in Canada. Canada's national assessment service does not itself publish a language requirement.

This is not registration advice. It is a reading of published CELBAN Centre pages, linked below.

How we researched this guide

We read the CELBAN Centre site through a browser rather than a command line fetch, then went to the speaking test page specifically rather than working from the landing page.

That step mattered. The landing page is marketing: it repeats that preparing for CELBAN is preparing for nursing, several times, and gives no format detail at all. Every fact in this guide comes from the dedicated speaking test page one level down.

The comparison table came from timings we had already extracted from published sources for other exams, so nothing there is estimated. Putting them together is what makes 35 minutes read as an outlier rather than as a number, and it also surfaced the more interesting point: every other long speaking test gets its length by adding candidates, and this one does not.

We have deliberately left the required score out. It is the obvious thing a reader wants and it is set provincially, so quoting a number we had not verified would have been the easiest way to make this guide wrong.

Practise speaking, from SpeakShark

SpeakShark is an AI English speaking practice app, and CELBAN is the closest published exam format to what it actually does. Role play, against a prompt you can see, with something responding to what you said.

You talk, the AI answers what you actually said, and you get speaking feedback while the conversation is still running. The free tier gives basic feedback; the detailed pronunciation and grammar breakdown is on Premium.

Being straight about the limits: a free session runs five minutes with four turns, which is daily practice rather than a 35 minute simulation, and we are a general speaking tool rather than a clinical English course. For nursing scenarios specifically you will want material built for that alongside this.

Start free with three sessions a day and no card. Paid sessions run ten minutes with unlimited turns. Limits are on the pricing page, and how it works walks through a session. Sign up here if role play is the part you want to stop dreading.

We are a speaking improvement tool. We are not an exam preparation provider, and we are not affiliated with Cambridge English, the IELTS partners, Pearson or any exam board. For test format, booking and official practice material, go to the exam body directly. Use SpeakShark to make your spoken English stronger, and use official material to learn the test.

• Site: speakshark.com
• Support: speaksharksupport@gmail.com
• Free tier: 3 AI conversation sessions per day, no card

Sources

• The CELBAN Centre, CELBAN Speaking Test: the up to 35 minutes figure, the Zoom delivery, the wired headset requirement, the five minute lateness rule, the mix of questions and role play, the prompts shared on screen and left visible for the duration of the task, and the recording for independent second rating
• The CELBAN Centre: the description of CELBAN as a nursing specific-purpose language exam assessing listening, writing, reading and speaking in the context of nursing
• National Nursing Assessment Service: the national assessment route for internationally educated nurses in Canada, which does not itself publish a language requirement
• Council of Europe, CEFR Table 1, common reference levels global scale: the general level scale referenced across this series

FAQ