Versant’s Assessment Tools Incorporate Artificial Intelligence
After the test, a test-taker’s response becomes a digitized representation of their speech. Versant’s patented speech technology sends this response to Versant’s servers, where proprietary speech software processes them, extracting bytes of information. The technology analyzes every utterance in incredible amounts of detail, specifically how the test-takers speak, as well as their manner of their speech.
vocabulary and language
The technology also analyzes how the test-takers respond to the questions, and how they pronounce the specified words. In addition, the software determines how well the applicant is able to answer questions in grammatically-correct sequences, using the appropriate lexicon and syntax.
The data is fed into two vital sub-scores: Content and Manner of Delivery, which measure language proficiency levels.
content and manner of delivery
The technology measures the two main manners of speech:
- Content, which covers sentence mastery and vocabulary, and
- Manner of Delivery, which covers fluency and pronunciation
Patented speech recognizes measure the test-taker’s utterance, then evaluate how long it took the test-taker to respond, the rate of speech and how long they paused, as well as their correct stress in words and phrases, and correct pronunciation, enunciation and intonation.
The software measures how well the test-taker was able to speak English compared to a native English speaker, to make sure they’re clearly understood.
Automated Scoring is Free of Bias
Because the scoring of the Versant Assessment Test is automated, it’s also free of bias or fatigue that can often effect human judgment. This automated process is exceptionally fair and virtually error-free, which reflects accurate, reliable results.
Results can also be seen online in as little as 5 minutes.