Student
Record with a webcam and microphone
Finish four parts in about fifteen minutes. Read one clear band with criterion feedback.
SpeakNusa is an English speaking assessment taken in the browser. Record video answers to five timed prompts across four parts, then receive a one to six band from two independent human examiners.
Five steps carry every submission from the first device check to the final band.
Input level
Browser only
| Band | Pronunciation | Fluency | Vocabulary | Grammar |
|---|---|---|---|---|
| 6 · Confident | Easily understood. Natural stress and intonation. Accent never impedes. | Sustained, natural pace. Hesitation only for content planning. | Wide, precise range. Natural collocation and effective paraphrase. | Full range of structures with consistent accuracy. |
| 5 · Competent | Generally clear. Occasional misplaced stress. Minimal listener effort. | Mostly smooth. Occasional repetition and self correction. | Sufficient range for all parts. Some precise word choice. | Mix of simple and complex structures. Frequent accuracy. |
| 4 · Functional | Understandable with some listener effort. Recurring sound and stress patterns. | Noticeable pausing. Turns sustained but unevenly. | Adequate for familiar topics. Struggles with abstract topics in Parts 3 and 4. | Limited complex structures. Errors occur but meaning is clear. |
| 3 · Developing | Frequently unclear. Pronunciation often obscures meaning. | Halting and fragmented. Long pauses. Hard to sustain a turn. | Narrow range. Frequent wrong word choice. Reliance on simple lexis. | Basic structures dominate. Frequent errors impede meaning at times. |
| 2 · Basic | Very difficult to understand. Heavy first language interference. | Very slow, with frequent breakdowns. Only short utterances. | Isolated words and phrases. Cannot paraphrase. Constant repetition. | Little control of basic forms. Meaning often obscured. |
| 1 · Minimal | Unintelligible or near silence. No assessable speech. | No connected speech beyond isolated words. | No productive vocabulary demonstrated. | No assessable grammatical control. |
Student
Finish four parts in about fifteen minutes. Read one clear band with criterion feedback.
Examiner
Play verified recordings, enter four criteria per answer, and complete a scoring that can no longer be edited.
Admin
Question bank and prompt audio, payment control and waivers, examiner assignment sets of two.
A browser, a webcam, and a working microphone in a quiet room. The device check shows your camera preview and microphone level before any prompt opens. Nothing to install.
About fifteen to twenty minutes, setup included. Each answer comes with its own preparation time before a timed recording window.
Part 1 Personal response (two short tasks, 1A and 1B), Part 2 Monologue from a cue card, Part 3 Decision-making between four options, and Part 4 Opinion. That is five recorded answers per submission, all from the same Test Set. Prompt text, timing, and prompt audio are snapshotted at delivery, so later question bank edits never rewrite a past session.
No. Like a live speaking exam, each answer is recorded once: recording starts when preparation ends and stops when speaking time ends. You can replay each question's audio once. The server verifies size, duration, and type before an answer counts toward the submission.
IDR 150,000 per assessment, paid through iPaymu. You only pay once your answers are recorded and verified.
Four criteria scored separately: pronunciation, fluency, vocabulary, and grammar. Each takes half band values from 1.0 to 6.0. The overall for one answer is the arithmetic mean of the four criteria.
Both work independently and never see each other's scores. Your report shows the combined band alongside both examiners' comments. A completed scoring is final and cannot be edited.
Verified recordings move to scoring once payment is settled or waived. Your submission moves from awaiting payment to paid, scoring, and scored, and your report and certificate open when scoring completes.
Only your assigned examiners and administrators, through short-lived signed playback links. Retention holds and purge requests govern how long evidence is kept.
The full session takes about twenty minutes, setup included.
Start your assessment