The official CELPIP score report gives a CELPIP level for the Speaking section as a whole, plus level descriptors for the four speaking dimensions, a Performance Profile of typical ability at that level, and Things to Work On suggestions.
All of that describes your overall Speaking level. It is not a set of scores for Tasks 1 through 8. CELPIP does not display individual task results, and it does not publish percentage weightings for the eight tasks — so you cannot add up your eight recordings using weightings you found online and call the total an official Speaking Overall score.
How the rating process works
According to official CELPIP score information, a single Speaking performance is typically evaluated independently by three to five trained Speaking raters. Raters do not know what the others have given.
Once rating is complete, the system checks whether the ratings are consistent. Where they diverge, an experienced benchmark rater may be brought in for further evaluation. The final result comes from performance across the scoring dimensions and is converted into a CELPIP level.
The four Speaking dimensions
- Content / Coherence: the number and quality of your ideas, how they are organised, and whether they carry examples and supporting detail.
- Vocabulary: word choice, accuracy, range, and whether the expression sounds natural.
- Listenability: rhythm, pronunciation, intonation, pauses, self-correction, grammar, and sentence structure — whether the answer is easy to follow.
- Task Fulfillment: whether the content is relevant and complete, whether the tone fits the situation, and whether the response is long enough.
All eight tasks provide evidence
Task 1 and Task 7 record for 90 seconds. The other recorded tasks are 60 seconds. Task 5 Part 1 only involves choosing an option and produces no recording; Part 2 is where the 60-second answer comes from.
Your official result will not tell you what any single task scored, and there is no published material showing that longer recordings automatically carry more weight.
Accent is not a separate criterion
Accent is not scored on its own. Official advice is not to worry about your accent unless it interferes with being understood. What is genuinely relevant is whether pronunciation, rhythm, and intonation — inside Listenability — make you easy for a rater to follow.
Speaking faster does not raise your level either. Too fast and sounds get swallowed; too slow and you have less content to develop. A normal, steady, clear pace matches real communication best.
Task Fulfillment is chronically underrated
When giving advice, speak directly to the person in the prompt. In a difficult situation, be considerate but firm. In the comparison task, genuinely persuade rather than merely compare. In the unusual-situation task, make someone who cannot see the picture understand the scene.
Even with good grammar and pronunciation, failing to do what the prompt asked or using a clearly unsuitable tone will limit your overall performance.
One weak task does not sink the section
A single pause, one grammar slip, or one task that did not go well does not mean the Speaking section has failed. CELPIP assesses the overall speaking performance formed by all eight tasks.
That is not a reason to give up on any task, though — every recording is another opportunity to demonstrate content, vocabulary, listenability, and task fulfillment.
How to self-assess honestly
You can analyse each practice recording against the four dimensions — but that is practice diagnosis, not an official per-task score.
The most useful record notes:
- whether the answer was complete;
- how many main ideas and supporting details you gave;
- which words were repetitive or imprecise;
- which pauses and pronunciations interfered with understanding;
- whether the tone matched the role.
Only after analysing several tasks in a row will the stable problems that actually limit your Speaking Overall become visible.
FAQs
How many raters? Typically three to five trained Speaking raters, working independently.
What if they disagree? An experienced benchmark rater may evaluate the performance further.
Do 90-second tasks count more? No published material says so, and no per-task weightings are released.
Next, read how Task 1 is scored or the Speaking format overview.