How the AI calculates the CEFR level
How the AI calculates the CEFR level
Designed by linguists, powered by AI, the 20-minutes Dynamic Speaking Test uses real-word tasks to deliver a detailed CEFR level. Before you take the Dynamic Speaking Test demo, it’s important that you take a moment to understand how the test works.
The Dynamic Speaking Test is a standardised assessment which measures language competence against the Common European Framework of Reference for Languages (CEFR). It uses a task-based, communicative approach to assess test takers' speaking abilities across all CEFR levels, catering to both novices and those at advanced C1-C2 levels.
The Dynamic Speaking Test uses AI to evaluate candidates based on their input against five key criteria:
Clarity of speech and proper intonation; making the speech intelligible.
Production of grammatically correct sentences.
Maintenance of a natural speech flow without excessive pauses.
Use of an appropriate and broad range of vocabulary.
Ability to handle various types of communicative tasks appropriately.
For more detail, download this pdf.
It's crucial to understand that the AI measures only the provided input. Unlike human raters, AI does not bring cultural bias into the scoring process, which can lead to inconsistent results influenced by factors like the test taker’s accent, gender, or other personal characteristics. The Dynamic Speaking Test is designed to be free from such biases, promoting fairness and consistency in scoring.