The Happiest Small Town: Preserving Hometown Memories of Kluang with an AI Music Video
What does your hometown look like in your memories?
Is it the outline of Gunung Lambak (Mount Lambak) shrouded in early morning mist, the traditional coffee shop (Kopitiam) on the corner filled with the aroma of charcoal-toasted bread and coffee, or the century-old railway tracks carrying countless stories of departures and returns?
For me, these scattered fragments of Kluang, Johor, Malaysia, are not just marks of growing up, but the deepest emotional anchor in my heart. Recently, I turned these sentimental memories into an AI music video project: 《最幸福的小镇》 The Happiest Small Town.
This project combines music, storytelling, and generative AI to capture the warmth of a Malaysian hometown, and serves as a heartfelt tribute to my hometown.
The Birth of the Idea: When Technology Meets Hometown Sentiment
Traditional music video production typically requires high budgets, collaborative teams, and tedious post-production, which makes it challenging for highly personal, localized cultural themes to be visualized. However, the rise of generative AI has presented us with new possibilities.
For this project, my goal was to establish a highly controllable, emotionally resonant AI-collaborative creative workflow that merges abstract nostalgia with concrete Kluang landmarks, creating an audio-visual piece with both cinematic quality and human warmth.
Behind the Scenes: The AI-Powered Music Video Workflow
This MV was not a simple "one-click generation," but rather a product of deep human-AI collaboration. Here is my core workflow:
1. Creative Direction & Lyrics (ChatGPT + Wesley Chong)
Lyrics are the soul of the song. Collaborating with ChatGPT, we first established several key visual motifs of Kluang:
- The lush greenery and high peaks of Gunung Lambak (Mount Lambak), representing the guardian of the town;
- The horn of the train and the tracks of the Kluang Railway Station, symbolizing departures and returns;
- The aroma of charcoal toast and rich coffee at a local Kopitiam, capturing the warmth of daily neighborhood interactions;
- Old streets, family, friendship, and the simple happiness of small-town life.
ChatGPT helped organize the structure of the verses and chorus, offering poetic sentence structures. I then localized and refined the lyrics manually to ensure that every word carried the unique charm of a Malaysian-Chinese small town.
2. Melody and Vocals (Ace Step 1.5 XL)
With the lyrics ready, I turned to the AI music generation model Ace Step 1.5 XL. By inputting custom prompts (such as Acoustic folk-pop, warm male vocals, nostalgic, emotional guitar progression) along with the bilingual lyrics, Ace Step 1.5 XL generated a melodious, warm, and comforting folk-style song. The chorus is extremely catchy, and the vocals are delicate and filled with narrative depth, perfectly capturing the nostalgic happiness of small-town life.
3. Visuals and Video Generation (LTX 2.3)
To bring the auditory memories to life, I utilized the latest open-source video generation model, LTX 2.3. I crafted detailed visual prompts for each lyrical scene to maintain style consistency:
- "Warm sunlight streaming into a wooden traditional Malaysian kopitiam, steam rising from cups of hot coffee, old friends sitting around laughing";
- "A vintage train slowly pulling into the lush green Kluang railway station, wildflowers blooming beside the tracks";
- "The grand silhouette of Gunung Lambak under a warm sunset, the sky painted in warm shades of golden and pinkish-orange."
LTX 2.3 performed exceptionally well in lighting, character consistency, and natural physical motion, giving the visuals a dreamlike, warm aesthetic reminiscent of a Studio Ghibli animation.
4. Directing and Post-Production (Wesley Chong)
Finally, I imported the generated audio and video clips into editing software, aligning the visual rhythm with the music, fine-tuning the colors, adding subtitles, and designing transitions. AI provided the high-quality puzzle pieces, while the human role was akin to a director, fitting those pieces together to tell a cohesive emotional story.
Beyond Technology: AI as a Creative Partner for Culture and Emotion
During the production of The Happiest Small Town, my greatest realization was that the value of AI lies not just in automation and efficiency, but in its ability to be a creative partner for expressing emotions and preserving culture.
Every small town has its unique stories, and every generation has its indelible memories. In the past, turning these memories into a beautifully produced music video required immense resources. Today, with the help of AI tools like Ace Step 1.5 XL and LTX 2.3, independent creators can capture hometown warmth, historic streets, and the bonds of friendship and family in moving visuals and melodies.
AI lowers the technical barrier but amplifies the warmth of humanity.
Listen and Connect
The Happiest Small Town is not just a story of Kluang people, but a reflection of anyone who grew up in a small town and is now striving in the big city.
What are your thoughts on this AI-powered creative workflow? Are you also exploring AI for music and video production? Especially to fellow Kluang people, I would love to hear your thoughts and memories!
Let's use technology to record warmth and AI to tell our own hometown stories.
