Culture & Arts
Wheat ears, fruits and vegetables become a stage for tiny workers, making the joy of harvest work songs visible.
Click an image to view it full size
01 / Background
“Songs of the Harvest” had to turn a cheerful work song into a harvest world you can see: sorghum, millet, corn and potatoes all had to appear, and the spirit of lines like “put your backs into it,” “the hard work pays off” and “carts piled high” had to build with the images. The content falls into three parts: the call to work, the harvest of all kinds of produce, and machine harvesting and transport, so the song isn't paired with a pile of unrelated farm-produce pictures.
The film uses miniature scenes, putting tiny farmers in the same frame as giant crops. Corn becomes a worksite to harvest and haul, a ladder leans against the grapes, and people rest in the shade under rice ears. With the change in scale, the crops still look familiar, but the work gains a fairy-tale charm.
The song lists one kind of produce after another, so the images needed both variety and a clear rhythm. We widened the differences in shot size, made the red of the sorghum stand out, gave persimmons, crabs, cotton and melons a four-panel grid, and gave the drones and harvesters a three-panel split. The film cuts between close-ups, groups of miniature figures and composite frames, and ends with a fully loaded convoy to show the result of the “harvest.”
02 / Creative Highlights
The film has tiny figures harvesting corn, hauling potatoes and climbing ladders by the grapes, rather than simply placing them next to crops as decoration. Giant produce and clear actions together establish the scale, so viewers see at a glance that “the harvest is underway.”
The lion dance, running through the fields and raised hands in celebration at the opening match the work song's high spirits; close-ups of crops match specific lyrics. Shot sizes and actions alternate between strong and soft, and each time the song's recurring beat comes back it gets a different image, so not every line has to be explained by subtitles.
Persimmons, crabs, cotton and melons appear together in a four-panel grid; the machinery section places drones and harvesters in the same frame. Combined, they show many kinds of produce in a short time and scale up each round of work into a busy, large-scale harvest.
The final cover spells out the five-character Chinese title in grain and straw. The warm golden texture, shallow depth of field and rice ears in the background carry on the film's miniature feel, so the title alone tells you the theme is the harvest.
03 / Scene Breakdown
A lion dance lifts the mood, and the camera then drops into the fields: tiny figures bend to cut rice, with a close-up of red sorghum and millet ears for each line, so the song's “put your backs into it” becomes the work right in front of you.
On a miniature ground shot from above, a golden lion dances in the middle of a ring of farmers, surrounded by piles of orange fruit and straw, as the “he-hai” subtitles pop up on the beat.
In backlit hillside paddies, rows of tiny figures in bamboo hats bend to harvest, matching the lyric “Put your backs into it, ya-mo-he-hei.”
First a close-up of a bright red sorghum head, then a cut to tiny figures busy harvesting beneath giant millet ears, with the vocable “si-li-li-li” turned into bouncing animated text.
Children run through the rice, the camera suddenly pulls back to a valley where dozens of tiny figures raise their hands in celebration, then pushes in on mountains of fruits and vegetables, turning “a bumper harvest” into a lively scene.
Three children run along a field ridge into the rice, then the view opens onto distant mountains and terraces, with crowds of tiny figures waving and dancing in the rolling wheat.
Tiny figures in the rolling wheat keep harvesting, then the shot cuts to small mountains of grapes, apples and corn, where a few figures sit and rest.
Whatever crop the lyrics name, the images show it: tiny figures climb onto ears of corn and fill baskets in a pile of potatoes, then a four-panel grid lays out persimmons, crabs, cotton and honeydew melons in one go.
Ears of corn as big as small hills, with tiny figures standing on them husking corn and passing tools; the scale is clear at a glance.
Tiny figures gather around potatoes bigger than themselves, hauling them and filling baskets, matching “Potatoes, ya-mo-hu-hai.”
Four matching bamboo baskets of persimmons, crabs, cotton and honeydew melons are combined in one frame; then the figures in the field slowly dissolve into harvesters at work and tiny figures carrying persimmons.
This melody is the longest and the images are the densest: persimmons, cabbages, grain piles, chili peppers, grapes and walnuts appear in turn, with a shot of tiny figures resting in the shade under big rice ears in the middle, and a six-panel grid to close.
Tiny figures dance hand in hand among orange-yellow fruit covering the ground, followed by a persimmon tree heavy with red fruit, with people gathering fruit beneath it.
People harvest next to a giant cabbage, a thatched hut steams beside a grain pile, and then a few tiny figures in big bamboo hats rest in the shade under rice ears as grains drift through the air.
A field of red chili peppers, climbing a ladder to pick grapes and an orchard where walnuts are knocked down flash by in turn, before six scenes come together and the pace tightens at once.
The colors follow the lyrics: first red sorghum in the setting sun, then drooping golden rice ears; the line “driving the harvester” uses a three-panel split of drones and harvesters to bring modern farm machinery into the song.
Backlit sorghum heads glow red, and the shot cuts to rice ears bowed under their weight; the two colors line up exactly with the two lyrics.
Three stacked panels show drones patrolling the fields and harvesters at work at the same time, then cut to a close-up of a harvester kicking up wheat dust, set to “wu-tu-tu-tu.”
The harvest is loaded up: tractors pull fruits and vegetables across the field ridges, a long line of fully loaded trucks rolls in, and finally dozens of shots come together in one big harvest mosaic.
On a path through a tomato field, tractors and small trucks drive out one after another, their beds piled with red fruit.
An overhead shot of a truck convoy loaded with grain, melons and fruit, then all the film's images shrink into one big mosaic, closing on “carts piled high.”
04 / Storyboards & Final Frames
Arranged in the order of the final film: each shot first notes which lyric it matches and what it shows, followed by the final image used in the film.
Shot 1.1
“Come on, folks, ya-mo-he-hai”: a miniature ground shot from above, a lion dancing among farmers, surrounded by piles of crops.
Shot 1.2
“Put your backs into it, ya-mo-he-hei”: rice paddies on a hillside at sunset, tiny figures bending to harvest.
Shot 1.3
“Sorghum and millet”: backlit sorghum heads in rich red, instantly recognizable as sorghum.
Shot 1.4
“Si-li-li-li”: giant millet ears weigh down the left of the frame as tiny figures work beneath them.
Shot 2.1
A few children run along a field ridge into the rice, which stands taller than they are.
Shot 2.2
“A bumper harvest, ya-mo”: distant mountains and terraces in the background, crowds of tiny figures dancing with raised hands.
Shot 2.3
“Hu-hai”: in the dense rolling wheat, rows of figures harvest their way forward.
Shot 2.4
Grapes, apples and corn piled into small mountains, with tiny figures sitting on top for a rest.
Shot 3.1
“Big corn, ya-mo”: tiny figures working on top of ears of corn.
Shot 3.2
“Potatoes, ya-mo”: tiny figures gather around giant potatoes, filling baskets.
Shot 3.3
Persimmons, crabs, cotton and honeydew melons all go in matching bamboo baskets, so the four panels combine into one frame without looking cluttered.
Shot 4.1
Tiny figures dance hand in hand among orange-yellow fruit covering the ground.
Shot 4.2
Beneath a canopy heavy with red fruit, tiny figures gather persimmons.
Shot 4.3
A giant cabbage stands in the center, with people harvesting around it.
Shot 4.4
Easter egg shot: a few tiny figures in bamboo hats rest in the shade under giant rice ears.
Shot 4.5
Red chili plants grow taller than people, with two tiny figures picking in the field.
Shot 4.6
A yellow ladder leans against dew-covered purple grapes.
Shot 4.7
A bamboo basket piled with walnuts, as tiny figures knock fruit down with bamboo poles.
Shot 4.8
Sunlight filters through the leaves as people in the grove harvest fruit.
Shot 5.1
“Red sorghum, ya-mo-hu-hai”: backlit sorghum heads.
Shot 5.2
“Golden rice, ya-mo-hu-hai”: drooping golden rice ears.
Shot 5.3
“Driving the harvester”: drones patrolling fields and harvesters at work, stacked in three panels.
Shot 5.4
“Wu-tu-tu-tu”: a harvester rolls toward the camera through wheat dust.
Shot 6.1
In a tomato field, tractors drive out pulling loads of fruits and vegetables.
Shot 6.2
“Carts piled high, ya-mo-hu-hai”: a long line of trucks loaded with grain, melons and fruit.
05 / Results
Completed a roughly 50-second landscape music short and a matching title cover, covering farmers at work, many kinds of autumn produce, machine harvesting and fully loaded transport. Released by CCTV.com on Kuaishou in late September 2024, it passed 6.48M views on Kuaishou.
Case Study
During the 2024 autumn harvest, CCTV.com released a 50-second AI music short, “Songs of the Harvest,” on Kuaishou.
The song was assigned to us, with one crop per line and strings of vocables like “si-li-li-li” and “suo-luo-luo-tai” in between.
We put the whole song into a miniature world: tiny figures working among giant corn, potatoes and rice ears.
The shots were generated as video with Kling AI, and after release it passed 6.48M views on Kuaishou.
01
The song was officially assigned and couldn't be changed. Nearly every line names a crop: sorghum, millet, corn, potatoes, persimmons, crabs, plus long stretches of vocables.
If each line were matched with any old picture of farm produce, the film would look like a field guide. We first broke the whole song into a storyboard, line by line: for each line, we wrote down what to show, the mood, wide shot or close-up, the colors and whether to use a split frame, then generated and selected images against that sheet.
“Come on, folks, ya-mo-he-hai” is matched with a lion dance and “put your backs into it” with harvesting in the fields, so the very first line puts people and work on screen.
02
Rural scenery has been filmed plenty of times on this theme. The client also asked that the images not look like existing harvest videos. We chose a miniature landscape: people made tiny, crops made huge.
The key was getting the tiny figures to really work: climbing ears of corn to husk them, hugging potatoes into baskets, setting up ladders to pick grapes. The crops look as familiar as ever, but once the actions are added, viewers see at a glance that “the harvest is underway,” with a touch of fairy tale too.
03
“Si-li-li-li si-la-la-la suo-luo-luo-tai” appears twice in the song, with no specific crop to draw. The client wanted dissolve transitions in the film, so we turned these two spots into gradual dissolves: the figures working in the field slowly melt away into a field of harvesters at work, then into tiny figures carrying persimmons.
Short shouts like “he-hai” and “hu-hai” became animated text bouncing on the beat, so viewers can both hear and see them.
A four-panel grid first lays out the autumn produce, then a dissolve carries the images from harvesting to hauling.
04
The line “the autumn harvest” lasts only a second or two, yet we wanted to show persimmons, crabs, cotton and honeydew melons at once. We made it a four-panel grid, and put all four in matching bamboo baskets so the combined frame looks neat rather than cluttered.
The line “driving the harvester” uses a three-panel landscape split: drones patrolling the fields above, harvesters at work below. The traditional work song now features modern farm machinery too.
All three panels get to work at once, followed by a harvester close-up set to “wu-tu-tu-tu.”
05
Halfway through, it struck us that a harvest film could fulfill the “dream of resting in the shade under the rice plants.” So we added a shot: a few tiny figures in bamboo hats rest in the shade under rice ears taller than they are, with grains drifting in the sunlight.
This shot sits in the middle of the dense work in the “hard work pays off” section, giving viewers a moment to breathe.
06
For the cover, we tried lettering made from mixed crops and from corn kernels, and the client finally chose the rice-ear version: the five characters are built from grain and straw and stand in a rice paddy, matching the film's miniature feel. The film's opening also uses this image as its title card.
07
For the client, an assigned work song became a short with memorable images: it passed 6.4M views on Kuaishou within three days of release, eventually exceeding 6.48M.
For fellow AI video creators and students, the project shows one thing: break the lyrics into a storyboard line by line first, then set a unified visual style, and AI image generation and image-to-video have a clear direction, so dozens of shots still feel like one film.
FAQ
This film consistently uses a miniature landscape and warm golden backlight: small people, big crops, blurred backgrounds. Each lyric was storyboarded first, images were generated in the same style, and Kling AI then turned the final images into video, so the thirty-plus shots share a consistent style.
The two “si-li-li-li si-la-la-la” passages became dissolve transitions, and “he-hai” and “hu-hai” became animated text bouncing on the beat. The vocables carry the rhythm, while specific produce is left to the lines with actual lyrics.
Use split frames. “The autumn harvest” uses a four-panel grid to show persimmons, crabs, cotton and honeydew melons at once, all in matching bamboo baskets; “driving the harvester” uses a three-panel landscape split for the drones and harvesters.
This project delivered a roughly 50-second landscape film at 1920×1080 and a matching title cover. Released by CCTV.com on Kuaishou in late September 2024, it passed 6.48M views on Kuaishou.
Approach
While refining the storyboard, we set the content, mood, shot size, color and split-frame approach for each lyric; during image selection, we added shots based on the color of the sorghum, the range of produce and the shot distance. Kling AI then turned the miniature work scenes into video, arranging close-ups, wide shots and multi-panel combinations within one unified warm-toned world.
The title design works the shape of rice ears into legible Chinese characters, echoing the harvest theme with the texture of crops. The warm-toned miniature world of work and the rice-ear lettering give the song an instantly recognizable look.
More Work
Ideal for brand launches · public-sector communications · cultural-tourism campaigns · IP transformation · event openers · earnings visualization · social campaigns · AI workflow development
