🪩 Atmos Night in Spherical Harmonics
Atmos Night
This song, “Atmos Night”, was created through a joint effort between myself, GPT, and SUNO, written for personal reflection and as a small birthday project.
The song reflects many late nights similar to those described in the lyrics, though not always focused on music itself. Glowing screens, AI-assisted coding, guitars resting in the room, and getting through the night with a can of Monster. It’s less a narrative than a snapshot: a brief record of learning and quiet focus.
lyrics:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
[Verse 1]
A can of Monster, cold and still
On a desk in Midtown Atlanta
Windows glowing soft against the sky
Spherical harmonics spin on the screen
Pulling sound into the monitors
He talks to the cursor
Line by line
[Verse 2]
He’s mixing dreams in Atmos
Rhythm drifts, the melody dances
Reverb echoes reaching in from afar
Every note takes its own track to open up the space
[Chorus]
Starlight overhead, lakalaka, slow and bright
Drumbeats flooding the ground, boom boom alive
The Les Paul soaring at nine o’clock in the sky
While the Taylor on the right gently weeps
Humming close, soft and slow
whisper the secret only he knows
[Guitar solo]
[Outro]
The night weighs heavy on his eyes
Every dream he never named
Still flickers inside the sound he made
Whisper a secret only he knows
…only he knows
Ambisonic Viewer
The Ambisonic Viewer has been added to the Sound Lab collection. It is a browser-first, real-time visualization for spatial audio on the sphere, designed to run entirely on the client without backend computation. To achieve interactive performance in the browser, the viewer uses Gaussian density kernels as a lightweight perceptual approximation of spherical harmonic bases, providing an intuitive, non-exact representation of Atmos-to-Ambisonic spatial mapping. The focus is on clarity and spatial intuition rather than physically exact HOA reconstruction, making it suitable for inspection, exploration, and demonstration of spatial audio behavior in real time.