For the oral section, use the "Listening" feature to play a short stimulus dialogue. Then, pause the audio and "shadow" (repeat immediately) what the speaker said, mimicking their intonation, stress, and pause. This is the fastest way to sound natural in English.
