Assessed on content only
No facial, emotion, accent or honesty analysis. Confidence, fluency and appearance are never treated as evidence.
VeriViva looks only at what a person says about the procedure. It never infers competence, intent or character from how someone speaks, looks or presents, and it never analyses emotion or honesty.
Told before, not after
People are told what the viva records and why, and consent before it starts. They can withdraw consent later.
Consent is asked for in the product before a viva begins, not buried in a policy. Before finishing, participants check their answers, mark anything that isn't what they said and can add a note. Reviewers see both.
Grounded in your sources
Every question set records the exact file it was made from, and each question shows the passage it is based on.
Questions and expected evidence come from your own approved documents. VeriViva doesn't substitute generic guidance, and it doesn't state legal requirements on your behalf.
Accessible by design
An assessor-led alternative route and reasonable adjustments, so nobody is marked down for how they communicate.
Anyone who can't be on camera can have an assessor-led session, where the assessor records typed responses with them. Adjustments are part of the product, so nobody is penalised for how they take part.
Proportionate capture
Capture only what the purpose needs: the recorded answers, their transcript and the review.
Consent is given once, before the viva starts, and can be withdrawn. Role-based access limits who can see the evidence, and access is recorded in the audit trail.
Traceable end to end
Sources, questions, answers, reviewers, decisions and changes form one record you can stand behind.
Every viva produces an evidence pack: the transcript, the sign-off and the recording, versioned and fingerprinted. Decisions and changes, including any change a reviewer makes to a suggested measure, sit in the audit trail.