examples.py

"""Example shot prompts, copied verbatim from the authors' own
``prompts/test_*.json`` in https://github.com/jd-opensource/JoyAI-Echo.

Only the first two shots of each story are used, to keep a demo run short.
"""

VLOG_1 = (
    "ID_A is a young woman in her twenties with shoulder-length dark brown hair, a slim "
    "build, and a soft natural look. ID_A wears a loose light beige knit top and relaxed dark "
    "trousers, creating a quiet home-lifestyle appearance. ID_A’s voice is a clear young "
    "female voice with a soft mid-high register, gentle breathiness, and intimate vlog-style "
    "pacing. The mouth movement is clearly visible in the frame and remains naturally "
    "synchronized with the spoken line, especially on slower reflective phrases and lightly "
    "smiled vowel shapes. At normal speed, ID_A steps into the center of the frame, settles "
    "the posture, and begins speaking with a calm and thoughtful expression. The lip movement "
    "aligns closely with the audio throughout the full sentence, especially on the slower "
    "emotional beats. In a soft young female voice with reflective warmth, ID_A says, \"I did "
    "not plan to record tonight, but something about this room felt different when I walked "
    "back in.\" The shot uses realistic indoor vlog-style imagery with soft practical light, "
    "neutral tones, and a warm personal atmosphere. The tone feels calm, introspective, and "
    "naturally domestic rather than performative. A stable medium shot frames ID_A from the "
    "waist up, keeping the face, upper-body posture, and visible lip movement clearly "
    "readable while preserving some of the room behind. The background includes a white "
    "curtain, soft string lights, part of a small table, and the warm interior of a well-kept "
    "room. The setting feels personal, lived-in, and quiet. Very soft indoor room tone, light "
    "fabric movement, and subtle foot placement are audible. No prominent background music. "
    "The audio focus remains on speech and subtle room ambience."
)

VLOG_2 = (
    "ID_A is a young woman in her twenties with shoulder-length dark brown hair, soft natural "
    "makeup, and a gentle but observant expression. ID_A wears a loose light beige knit top "
    "and dark relaxed trousers. ID_A’s voice is a youthful female voice with clear diction, a "
    "soft intimate tone, and slow personal-story pacing. The mouth movement remains visible "
    "and synchronized as ID_A turns slightly toward the window side of the room while "
    "continuing to talk. At normal speed, ID_A turns slightly to the side and glances toward "
    "the curtain as if the room itself has triggered the reflection. The spoken line remains "
    "clearly mouth-synced, including subtle lip compression at clause endings. In a soft, "
    "thoughtful female voice, ID_A says, \"Maybe it is because I have been too busy to really "
    "look at any of this for a while.\" The visual style remains realistic and soft, with "
    "warm-neutral indoor tones and a personal diary-video mood. The emotional atmosphere is "
    "quiet and self-aware. A slight pan follows ID_A as the subject turns toward the curtain, "
    "keeping facial expression and mouth articulation visible while revealing more of the "
    "side wall. The curtain, a faint edge of the window area, and small warm decorative "
    "lights remain visible. The room continues to feel intimate and uncluttered. Light fabric "
    "movement, a soft shift of feet, and low indoor ambience are audible. No background "
    "music. The room’s quiet and the speaker’s voice dominate."
)

WAREHOUSE_1 = (
    "ID_A is a warehouse worker seated on a low metal bench, is a man in his thirties with "
    "short black hair, tan skin, a sturdy medium build, and a practical tired face. He wears "
    "a maroon T-shirt, grey work shorts, and is pulling on tall white rubber boots for "
    "wet-floor industrial work. ID_A’s voice is a low, steady male voice with a calm "
    "working-class plainness, slightly rough from long hours and warehouse dust, with natural "
    "conversational pacing. His mouth is visible and synchronized with the line. At normal "
    "speed, ID_A braces one hand on the bench and pulls the rubber boot firmly over his heel, "
    "glancing down to check the fit before looking ahead. His lip movement aligns with the "
    "line. ID_A says, \"If the floor is wet again tonight, I want to know before it gets "
    "halfway across the building.\" Naturalistic industrial drama with realistic warehouse "
    "lighting, practical textures, and documentary-like physical detail. A medium-close "
    "static shot focuses on ID_A from the knees up, keeping the boot-pulling action and his "
    "face in view. A worn warehouse prep area with concrete floor, metal lockers, stacked "
    "plastic crates, and a hanging fluorescent light casting a cold industrial glow. Rubber "
    "boot squeak, fabric friction, distant forklift reverse beep, fluorescent hum, and "
    "far-off metal clatter are audible. No prominent background music, only a faint low "
    "industrial tension pad underneath the natural ambience."
)

WAREHOUSE_2 = (
    "ID_A is a warehouse worker adjusting the second white rubber boot, is a man in his "
    "thirties with short black hair, tan skin, and a practical, focused expression. He wears "
    "a maroon T-shirt and grey shorts, with both tall white rubber boots nearly secured. "
    "ID_A’s voice is a low male voice with dry humor, steady breath, and an unhurried "
    "practical rhythm. His mouth is visible and synchronized with speech. At normal speed, "
    "ID_A presses the heel down, smooths the top of the boot with one hand, and gives it a "
    "final tug. His mouth movement stays aligned with the line. ID_A says, \"Nobody notices "
    "proper boots until the place starts flooding and everyone suddenly wants a pair.\" "
    "Realistic industrial workplace footage with tactile detail and grounded working-life "
    "atmosphere. A lower medium shot centered on the bench, boots, and torso, keeping facial "
    "reaction visible. Metal bench, industrial lockers, scuffed concrete, and stacked storage "
    "bins remain visible behind him. Rubber flex, bench creak, distant pallet drag, "
    "fluorescent buzz, and soft ventilation rumble are audible. No distinct score, just low "
    "atmospheric tension under the room tone."
)

RALLY_1 = (
    "ID_A is an American male racing driver in his thirties with a lean athletic build, sharp "
    "cheekbones, short black hair, light stubble, and a weathered face marked by experience "
    "and pressure. ID_A wears a partially unzipped professional rally racing suit over a dark "
    "fireproof base layer, with a serious focused expression before entering the car. ID_A’s "
    "voice is a low, steady male baritone with a rough edge, controlled breathing, and "
    "restrained emotional force. The face is shown clearly in close framing, and the mouth "
    "movement is highly visible and tightly synchronized with the spoken line. At normal "
    "speed, ID_A looks past the camera toward the road, then lowers his gaze for a brief "
    "controlled breath before speaking. Lip, jaw, and breath timing remain closely aligned "
    "with the line. In a low, steady male voice, ID_A says, \"I did not come back for a clean "
    "drive. I came back because this mountain still owes me an answer.\" Realistic motorsport "
    "film language with cool daylight, hard facial detail, and restrained cinematic tension. "
    "The tone is serious, professional, and emotionally loaded rather than flashy. Stable "
    "cinematic close-up centered on ID_A’s face and upper shoulders for strong identity "
    "anchoring and clear lip-sync. The background shows a mountain rally start zone with "
    "safety barriers, blurred crew members, a dust-coated race car, open air, and a winding "
    "road disappearing into pale rocky terrain. Engine idling in the background, faint wind "
    "over the mountain road, distant metallic clicks from the cooling car body, and soft "
    "fabric movement from the racing suit are audible. A very low cinematic tension pulse may "
    "be present, but it stays far beneath the spoken line and mechanical ambience."
)

RALLY_2 = (
    "ID_A is an American male racing driver in his thirties with short black hair, light "
    "stubble, and an intensely focused expression. ID_A wears a full rally suit with sponsor "
    "patches and tightens his gloves with practiced speed. ID_A’s voice is a firm low male "
    "voice with clipped diction and restrained intensity. His face remains visible enough for "
    "strong identity continuity, and the mouth movement stays naturally synchronized with the "
    "spoken line. At normal speed, ID_A tightens both gloves, glances once toward the "
    "driver-side door, and speaks without wasting motion. The mouth movement stays clearly "
    "synchronized. In a low controlled male voice, ID_A says, \"If I leave anything unfinished "
    "in the first section, the road will take it from me later.\" Realistic and tactile "
    "rally-film detail, emphasizing gloves, suit fabric, and body tension over broad "
    "spectacle. The mood is concentrated and severe. Medium-close shot following glove "
    "tightening and upper-body preparation while preserving facial readability and lip-sync. "
    "The race car fills part of the background, along with mountain haze, service tents, "
    "blurred crew, and safety markers at the start line. Glove strap tightening, zipper "
    "movement, engine idle, wind across metal surfaces, and a faint crunch of gravel "
    "underfoot are audible. A restrained low-frequency cinematic pulse remains secondary to "
    "the live motorsport soundscape."
)