A song about the sacred in daily life became a unified AI music video

Face of God Project Description Achievements The visuals aren’t random images paired with lines — they’re a single, unified visual meditation on the song.  Each image renders the lyric’s emotional and spiritual depth, and because they were generated as an integrated whole, the video holds together with consistency, resonance, and a clear inner thread. Tools ChatGPT Suno Gemini (Nano Banana)  Midjourney Airtable iMovie Methodology I view the Incarnation as the heart of the Christian tradition: the extraordinary is encountered in the ordinary, the sacred is revealed in the mundane, and the face of God is made known in daily life. I wanted to write a song that invites people to recognize the sacredness of life—and to notice how the Holy shows up in lived experience. I began by collaborating with ChatGPT to clarify the theological core of the project: that Christ reveals the face of God in humanity and, by extension, in all of life. From that conversation, we shaped the lyrics for The Face of God. Next, I moved to Suno, where I uploaded the lyrics, arranged the structure (chorus, bridge, verses), and selected the vocal style, genre, and overall sound. After several iterations, the final recording of The Face of God emerged. With the song complete, I envisioned a music video suitable for both the church website and Facebook—one in which the visuals would not merely illustrate the lyrics, but convey the depth and emotional tone of each phrase. To develop the visual language, I engaged Gemini as a creative producer and organizer. Gemini generated a detailed imagery breakdown in table form—mapping each lyric phrase to a visual concept and creative focus—which I exported into Google Sheets. From there, we defined the core visual elements (lighting, mood, theme, style), built a scene-by-scene plan with timing and intent, and created phased time cuts for verses, chorus, and bridge. With this structure in place, I generated the primary image set using Nano Banana, saving each prompt as I worked so I could refine and regenerate images as needed. I also used Midjourney to generate several additional images and to animate selected frames. These animated moments add movement and texture to the video, creating contrast between stillness and rush. Finally, I exported the Sheets into Airtable and recorded each image alongside its prompt and corresponding lyric. Airtable became the project’s organizational backbone: it ensured every phrase had an assigned visual and made it easy to locate the correct image during editing. When I assembled the final video in iMovie, Airtable streamlined the workflow and kept the visual sequence aligned with the music. (Airtable base: https://airtable.com/appeHF9Cg9hKCShJp/shramk1f5Wn3RjG6a) Once all images and animated clips were complete, I edited the final sequence in iMovie, aligning each visual with its lyric and musical phrasing to create a cohesive, resonant whole. Learnings My biggest takeaway was realizing that AI is most powerful when it helps shape the whole — not just the parts.  I would normally prompt for an image per phrase and then attempt to create unity in post.  But by engaging Gemini to support the project as a music video — essentially as a producer and editor — I created a consistent visual language across the entire piece (lighting, mood, theme, and style).  That shift is what made the final video feel coherent, complementary, and emotionally integrated. Relevant URLs https://airtable.com/appeHF9Cg9hKCShJp/shramk1f5Wn3RjG6a https://www.facebook.com/TriuneLutheran https://www.triunelutheran.com/the-face-of-god-1