# Block Noise Siren ## Subject / Scene Settings - Audience… | Deepfake AI Video

Text-to-video prompt used to generate this video

# Block Noise Siren ## Subject / Scene Settings - Audience: {locale="JP"}; Narrative tone: surreal horror; clinical - Subject type: humanoid singer - Key features: face fully masked by iridescent block-noise slab; lace white vintage dress; trembling hand; shallow breaths - Age: mid-20s; Vibe: eerie/fragile; Skin: pale matte; Makeup: minimal; Hair: concealed under noise - Outfit: high-neck lace blouse + gathered skirt + soft flats; Accessories: thin silver ring - Lighting: soft top-left key (frosted window); faint right rim; negative fill camera-left; minimal bounce; light haze {low}; thin vol shafts - Grade: Palette ice-blue/pearl white/graphite/iridescent RGB; gentle S-curve; light halation; subtle vignette; fine grain {low}; mild CA; clean flares - Visual taste: sterile grey void; micro exposure drift; practical window glow; real shadows/occlusion - Camera: MS→tight MS; centered→rule-of-thirds; sleeve occluder/parallax; ONE move: slow dolly-in (micro handheld jitter) - Lens/Focus: 50mm feel; shallow bokeh; brief gentle racks to hand/noise - Coverage: master+inserts; match-on-action; consistent screen direction; implied eyeline downcast - Persist: same white lace dress / same iridescent block-noise head throughout ## Audio (BGM & SFX) - BGM: dark ambient bed; bowed metal clusters; sparse sub pulses (45–55 BPM; long fades) - SFX: TV snow; digital glitches; buffer pops; breath; cloth creak; fingertip taps - Cues: swell on each touch/morph; tritone stab ~8s; final low-pass suck before blackout; mix_note: duck BGM by ~3 dB on collisions