AI changes you can use.

This is about checking whether speech-to-text is truly getting better, or just getting better at specific scorecards. The goal is to spot “test-chasing” so real-world performance does not disappoint.

The idea is to add two checkpoints that can say “no” to an AI-made draft. First, a teacher edits the script using known principles for how people learn from words and images. Second, computer checks look for confusing explanations and mismatches between what’s said and what’s shown.