- Document separate sentence-audio and word-pronunciation fields - Add timing review coverage for configured audio fields