Better flashcard architecture

The idea using Anki was to have a JP sentence on side A and the equivalent ENG on side B.

It doesn't work as an interpreting training tool because it suggests that the correct oral translation of A is B (and vice versa), whereas A vis-a-vis B are only one among possible interpretation of each others. The second approach then is to have a sentence on A and some keywords translated on B in order to fire on the switching process of oral translation among students. Adding the voiced over sentence is technically feasible but painful. On the other hand, I am favorably reconsidering using Vertov and its colleague Zotero over Firefox or an equivalent browser because it puts the oral or video on the center stage of the task.