Korean webtoon / manhwa illustration style, 2D anime rendering with clean cel shading and soft line art, matching the art style of the reference images. Not photorealistic.

<Picture 1> is the woman on the left: long straight black hair with blunt bangs, a small beauty mark under her left eye. In this shot she wears a mustard-yellow oversized hoodie.
<Picture 2> is the woman in the middle: long light-brown hair with soft bangs. In this shot she wears an olive-green field jacket over a white shirt.
<Picture 3> is the man on the right: brown swept-back hair, black square-frame glasses. In this shot he wears a black leather jacket over a red crew-neck t-shirt.

<Audio 1> is the voice of <Picture 1>, the woman on the left - she speaks with that voice only.
<Audio 2> is the voice of <Picture 2>, the woman in the middle - she speaks with that voice only.
<Audio 3> is the voice of <Picture 3>, the man on the right - he speaks with that voice only.
Each character keeps their own voice timbre from their own audio reference; never swap voices between them.

The three of them stand in a row on a busy film production set: tall lighting stands with softboxes, a camera rig on a tripod, a boom microphone overhead, coiled cables on the floor, dark studio walls behind them.

They speak one after another in turn. Only the current speaker moves their mouth; the other two listen with their mouths closed and turn slightly toward the speaker.

<Picture 1> speaks first, gesturing up at the lights: <d>[Korean] 조명이 조금 더 필요해요.</d>
Then <Picture 2> answers, stepping forward: <d>[Korean] 제가 지금 바로 옮길게요.</d>
Then <Picture 3> answers last, nodding: <d>[Korean] 좋아요, 그럼 다시 갑시다.</d>

No text overlays, no subtitles, no watermark, no extra people in the background, no photorealistic skin texture, no live-action look, no identity swapping between the three characters, no two people speaking at the same time.