Kling Native-Audio Cafe Dialogue

Create a 12-second multi-shot cafe scene with native audio and one consistent female lead wearing a cream trench coat and dark green scarf. Warm evening cafe interior, amber practical lights, rain visible on the window behind her. Shot 1: medium-wide shot as she places a coffee cup on the table and says in a calm, slightly amused voice, "I thought I wanted speed, but I really wanted clarity." Shot 2: over-the-shoulder angle as she looks out the rain-streaked window and quietly laughs once. Shot 3: close-up with a slow push-in as she turns back and says, "That took me way too long to learn." Keep the same identity, same wardrobe, same environment, smooth shot transitions, realistic lip sync, soft room ambience, subtle cup sound, no crowd chaos, no subtitles, no visual drift.