안녕하세요! 한지연입니다. 

자료실을 활용하여 AI 콘텐츠 제작 관련 내용 공유하겠습니다. 

첫번째 

Gemini을 통한 기본 사진을 만들어야 됩니다. 

배경은 회색 

강아지 나이, 품종, 색상을 얘기하고 나노바나나를 통해서 이미지를 생성합니다. 

두번째



나노바나나를 통해서 원하는 장면 원하는 이미지를 생성합니다. 

기본 이미지를 첨부하고 나노바나나에서 세이블, 포메라니안, 퍼피로 생성해서 영상용 이미지를 준비합니다. 

세번째

Gemini에서 인터뷰로 사용할 시간을 나누는 플로우용 프롬프트를 생성합니다. 

플로우는 길이가 8초기 때문에 8초에 대사가 적으면 천천히 말할 수 있고 대사가 많으면 급하게 말하게 됩니다. 

8초 이상은 잘리기 때문에 아래 참조 내용은 수정을 하셔야 됩니다. 

<참조>

[LOCK / REFERENCE]

Use the provided reference image as the single source of truth.

Keep Siru’s exact visual identity consistent: golden/orange fluffy Pomeranian-like baby dog, cream chest fur, round bright eyes, small black nose, smiling mouth with tiny tongue, same tuxedo (black jacket, white shirt, black bow tie), same guitar (sunburst electric guitar with white pickguard), same stage setting (wood floor, red curtain, orchestra blurred in background).

No subtitles, no on-screen text, no logos, no watermark.

[SCENE]

Animal interview on a concert hall stage. Siru stands center holding the electric guitar. Warm stage lighting. Background orchestra remains out of focus (bokeh), cinematic depth of field. 4K aesthetics, high-detail fur.

[CAMERA]

Vertical 9:16, close-up to medium-close framing (Siru from head to knees).

One continuous take, gentle handheld micro-movement, subtle slow dolly-in.

Focus locked on Siru’s eyes and mouth for clean lip-sync.

[MIC VISIBILITY RULE]

Reporter never appears (no face/body).

When the male reporter speaks: microphone hand is NOT visible (audio only).

When Siru speaks: a human hand holding a black handheld microphone enters from the right edge, points to Siru’s mouth area, then exits after Siru finishes.

[MOTION / GESTURES]

0.0–1.0s: Siru looks forward, then reacts to off-screen voice; slight head tilt, blink.

1.0–2.2s: Siru adjusts posture with the guitar (tiny paw shift), attentive expression.

2.2–5.6s: Mic hand enters; Siru speaks with clear mouth movement; playful gesture (small paw wave + cheeky grin).

5.6–7.5s: Siru finishes line and giggles innocently; reporter laughs off-screen; Siru smiles brightly.

[AUDIO / DIALOGUE (Korean, lip-synced)]

Siru (Korean 3-year-old boy voice, cute & mischievous, lip-synced): “두번째 곡 나야 볼레로가 나왔어요. 아주 심각한 음악이에요”

Male Reporter (off-screen, chuckle): “곡 작업이 많이 힘들었나요?”

Siru (Korean 3-year-old boy voice, cute & mischievous, lip-synced): “네 너무 힘들었어요”

Male Reporter (off-screen, chuckle): “곡이 많이 어려웠나요?”

Siru (Korean 3-year-old boy voice, cute & mischievous, lip-synced): “아니요 계속 반복하는게 너무 힘들었어요, 하하하하하하하”

Male Reporter (off-screen, chuckle): “그럼 새로 만든곡 들어볼까요?”

Natural laughter after the last line (reporter off-screen + Siru giggle).

[NEGATIVE / AVOID]

No subtitles/captions. No extra people entering frame. No reporter face/body.

No deformed hands/fingers, no extra limbs, no costume change, no scene change, no camera cuts.

Do not change Siru’s fur color, tuxedo details, guitar shape, or stage background.

[OUTPUT]

7–10 seconds, 24fps filmic motion, 4K look, high-detail fur, natural lighting, clean lip-sync.

(동영상 제작 금지)

네번째

구글 플로우에서 영상을 만들어주면 됩니다. 

https://labs.google/fx/ko/tools/flow

영상은 8초 단위로 계속 연결할 수 있습니다. 

마지막 프레임이 다음영상 첫 프레임으로 만들면 됩니다.