VdoBloom
← Back

Xianxia Comedy Echo Chamber Scene

Short Filmby @Soran · source ↗August 9, 2026

Full prompt

[Generation Mode]
Seedance
Generate a 0-10 second clip first, then extend by 5 seconds based on the clip and its final frame.
Total duration: 15 seconds.
[Screen Specifications]
16:9 Widescreen
Cinematic photorealistic quality
Native synchronized Mandarin dialogue
No subtitles
Always only two visible characters

[Overall Style]
Cinematic photorealistic texture, using monumental pure Chinese Xianxia architecture, restrained deadpan comedy, strong spatial audio narrative, warm morning sun penetrating cool blue-white sea of clouds, weathered megaliths, dark ancient building tile roofs, ivory silk, and fine film grain; the suspense ultimately resolves into a comedy reversal through the grand scene itself.
[Background Reference]
@Image1–@Image4 serve as background and location references only, never as character references.
Strictly preserve in the reference images:
Giant circular heavenly gate
Layered roofs of the celestial city above clouds
Mountain-scale stone pillars
Dense churning sea of clouds
Direction of the morning sun
Grand spatial depth
Incredible vertical scale

[Character Settings]
Character A | Sword Immortal Elder Sister
26-30 year old East Asian female sword immortal, oval fair face, narrow dark eyes, long black hair half-up secured with a white jade hairpin, tall and slender, wearing white embroidered silk Hanfu, semi-transparent layered wide sleeves, silver waist seal, jade pendant, white cloth boots, and a single slender silver flying sword.
Character B | Junior Sister
The same 20-24 year old East Asian junior sister, round and lively face, black hair in braids, petite build, wearing teal linen Hanfu, dark belt, wooden hairpin, and black cloth shoes.
[Camera Design]
Shot 1 | 0-5s | Wide Shot / Long Shot
24mm aerial shot slowly pushing toward the giant heavenly gate.
Character A stands on the same silver flying sword, hovering three meters above the stone platform, with the same junior sister standing behind.
The elder sister shouts deeply toward the giant gate:
"Heavenly Gate Guardian, reveal yourself!"

Inside the circular giant gate, a thick echo with clear spatial orientation immediately returns:
"Reveal yourself... self... self..."

Shot 2 | 5-10s | Medium Shot / Cowboy Shot
The same elder sister in white silk Hanfu lands back on the same stone platform, the same junior sister remains behind, background stays on the same giant heavenly gate and celestial city.
The elder sister narrows her eyes, half-drawing the same long sword, and says coldly:
"Do not hide your head and show your tail."

The giant gate responds again with a massive echo:
"Hide your head and show your tail... tail... tail."
Heroic guqin and sword hums gradually rise; she assumes a restrained and precise defensive stance.

Shot 3 | 10-15s | Close-up / Extreme Close-up
The final 5-second extension must continue exactly from the last frame of the previous 10-second clip, strictly maintaining:
Camera axis continuity
Character positions continuity
Sun direction continuity
Cloud movement continuity
Architectural structure continuity
Flying sword continuity
Clothing continuity

The same junior sister walks to her side and says calmly:
"Sister, that's just an echo."

Then she cups her hands to her mouth and shouts toward the gate:
"Sister cares about her reputation the most!"
The entire giant gate immediately booms in repetition:
"Reputation... tion... tion..."
The same elder sister slowly sheathes her sword and whispers:
"This gate is too talkative."
The gate immediately echoes back:
"Talkative... tive... tive..."

The camera stops on her face as she tries to maintain a dignified, profound look but stays completely frozen, with one eyebrow twitching slightly; the same junior sister turns her face away, silently stifling a laugh.

[Continuity Requirements]
The actual generation process is:
First, generate the full 0-10 seconds.
Then use this 0-10 second clip and its final frame as a continuity reference.
Only extend for the final 5 seconds.

[Technical Requirements]
Total duration 15 seconds
16:9 Widescreen
Native synchronized Mandarin dialogue
Giant gate echo has clear direction and long reverb
Realistic and natural movement of the cloud sea, silk fabric, hair, and flying sword
No subtitles
Always only two visible characters
[Negative Prompts]
blurry, bad quality, low quality, low resolution, noisy, jpeg artifacts, watermark, text, error; deformed, mutated, bad anatomy, poorly drawn hands, bad composition, out of frame, disfigured; inconsistent character, changing clothes, face morphing, background shift, glitching cuts, disappearing props

Original prompt

【生成模式】

Seedance
先生成 0–10 秒成片,再基于这段 0–10 秒成片及其最后一帧,连续延长后 5 秒。
总时长 15 秒。
【画面规格】
16:9 横屏
电影级写实质感
原生同步普通话对白
不生成字幕
始终只有两名可见人物

【整体风格】

电影级写实质感,采用纪念碑级纯中式仙侠建筑、克制的冷面喜剧、强烈空间声音叙事、温暖朝阳穿透清冷蓝白云海、风化巨石、深色古建瓦顶、象牙白真丝与细腻胶片颗粒;悬疑最终通过宏大场景本身完成喜剧反转。
【背景参考】
@图片1–@图片4 仅作为背景与地点参考,绝不作为人物参考。
严格保留参考图中的:
巨型圆形天门
层层叠叠的云上仙城屋頂

山岳尺度石柱

浓密翻涌云海

朝阳方向

宏大空间纵深
不可思议的垂直尺度

【角色设定】
角色 A|剑仙师姐
26–30 岁的东亚剑仙女子,椭圆白皙面庞,细长深色眼睛,黑色长发半挽,以白玉簪固定,身材高挑纤细,穿白色刺绣真丝汉服、半透明分层宽袖、银色腰封、玉佩、白色布靴,唯一一柄细长银色飞剑。
角色 B|小师妹
同一个 20–24 岁的东亚小师妹,圆润灵动的脸,黑发编辫,身形娇小,穿青绿色亚麻汉服、深色腰带、木簪和黑色布鞋。
【镜头设计】
镜头 1|0–5s|全景 / 远景
24mm 航拍缓慢推向巨大天门。
角色 A 踩着同一柄银色飞剑悬浮在石台上方三米处,同一个小师妹站在后方。
剑仙师姐朝巨大天门沉声喝道:
“天门守将,现身!”

圆形巨门内部立刻传回浑厚而具有明确空间方位的回声:

“现身——身——”

镜头 2|5–10s|中景 / 牛仔景

同一个穿白色真丝汉服的剑仙师姐落回同一座石台,同一个小师妹仍然保持在后方,背景始终保持同一座巨型天门与同一片云上仙城。
剑仙师姐眯起双眼,将同一柄长剑拔出一半,冷声说道:
“莫藏头露尾。”

巨门再次以巨大回声回应:
“藏头露尾——尾——。”
英雄式古琴与剑鸣逐渐升起,她摆出克制而精准的戒备姿势。

镜头 3|10–15s|特写 / 极近特写

后 5 秒延长必须从前 10 秒成片的最后一帧原位继续,严格保持:

摄影机轴线完全连续

两名角色位置完全连续

太阳方向完全连续
云层运动完全连续

建筑结构完全连续

飞剑完全连续

服装完全连续

同一个小师妹走到她身旁,平静说道:

“师姐,那只是回声。”

随后双手拢在嘴边,对着天门大喊:

“师姐最爱面子!”
整座巨大天门立刻轰然重复:
“爱面子——子——。”
同一个剑仙师姐缓慢将长剑归鞘,低声说道:
“此门,多嘴。”
巨门马上再次回荡:
“多嘴——嘴——。”

镜头停在她努力维持端庄、高深却彻底僵住的脸上,一侧眉毛轻轻抽动;同一个小师妹转开脸,默默憋笑。

【连续性要求】

实际生成流程为:

先完整生成 0–10 秒

再把这段 0–10 秒成片及其最后一帧作为连续性参考

只延长后 5 秒

【技术要求】

总时长 15 秒

16:9 横屏

原生同步普通话对白

巨门回声具有明显方向与长混响

云海、真丝衣料、头发和飞剑运动真实自然

不生成字幕

始终只有两名可见人物
【负面词】
blurry, bad quality, low quality, low resolution, noisy, jpeg artifacts, watermark, text, error; deformed, mutated, bad anatomy, poorly drawn hands, bad composition, out of frame, disfigured; inconsistent character, changing clothes, face morphing, background shift, glitching cuts, disappearing props
Generate this with Seedance 2 on VdoBloom.
Opens the image-to-video creator with this prompt and Seedance 2 pre-selected.
✨ Generate Video