The Solar Oracle Walkman

• 

  Sony WM-F107 exhibited in Solar Biennale 2025 in Lausanne.

• 

  The DIY I-V tester made by Marc Dusseiller.

Abstract

This multidisciplinary research combines DIY solar practice, conceptual music, and the exploration of AI latent space. A “solar oracle walkman” is prototyped that turns a handmade dye-sensitized solar cell (DSSC) into music. Central to the design is a 7-dimensional I–V “fingerprint”: the unique curve of each handmade DSSC is treated as a musical score, translating its photovoltaic character into reproducible, generative sound. Reinvented from a DIY I–V tester, the device is self-powered: a 6×6 cm “solar mini disc” can be inserted and rendered to audio in real time. Machine learning provides the medium for this translation. The sound engine is implemented in Max/MSP with RAVE, a neural audio synthesizer whose latent space encodes timbral variation. At present, the I–V fingerprint is manually mapped to the decoder’s latent inlets, a minimal design that prioritizes real-time responsiveness and artistic control. Beyond sculptural experimentation, the work also opens a research path: perception and AI can be viewed as two sides of the same generative mechanism, where the brain actively predicts and corrects sensory input, and hallucination is an extreme case of prediction mismatch. This project does not attempt to solve these questions now; the solar oracle walkman remains a simple art sculpture. The theories and the initial hypotheses mainly guide the design of datasets and priors for RAVE mapping and training. The project also fosters systematic collection of natural-dyed DSSC data and invites a discussion of whether solar energy could be made traceable.

• 

  The measurement of the I-V curve tester is uploaded to Thingspeak and a local server, and can be fetched in Max/MSP.

• 

  DIY DSSC with screen printed pattern and hollyhock dye made by Shih Wei Chieh.

• 

  DIY DSSC with cyanotype pattern made by Shih Wei Chieh.

</gallery>

RAVE as Latent Oracle: Design Flow and Philosophical Mechanism

This work proposes a RAVE latent oracle, where the latent space of a real-time audio variational auto-encoder (RAVE) serves as an interface between physical energy and machine intelligence. In blockchain systems, oracles provide access to external data that a closed ledger cannot sense. Likewise, most AI models are closed symbolic systems without direct perception. Dye-sensitized solar cell (DSSC) I–V fingerprints are introduced as inputs and processed by RAVE, functioning as a fuzzy extractor that maps noisy physical signals into reproducible latent coordinates. These embeddings act as an oracle that grounds the model with the uniqueness of energy curves. Philosophically, perception and AI can be regarded as two sides of the same generative mechanism. Brains do not passively receive the world; they predict and correct it. Hallucination is an extreme case of prediction mismatch. In this system, the oracle has a dual role: as an anchor, it constrains generation with empirical data; as an invocation, it allows imagination to survive under verification, keeping creative variance while containing drift. Stinson’s model of generic mechanisms clarifies that both neural systems and computational models can instantiate the same kind of generative architecture. The DSSC–RAVE pipeline is not a superficial imitation but another instantiation that transforms energy inputs into perceptual outputs, illustrating how controlled hallucination becomes both a technical resource and an artistic metaphor.

• 

  Diagram of Stinson's generic mechanism.

• 

  A logical empiricist picture of a scientific theory. Reproduced from Herbert Feigl 1970.

                  [ Generic Mechanism Kind ]
    (Grounded inference: sense ⇄ abstract ⇄ validate ⇄ act)
                               │
        ┌──────────────────────┴───────────────────────┐
        │                                              │
[ Solar Oracle System ]                       [ Human Cognition ]
------------------------                      ------------------------
Soil of sunlight & matter                     Soil of observation
         │                                              │
         ▼                                              ▼
DSSC I–V fingerprints (7-D)                   Empirical concepts
         │                                              │
         ▼                                              ▼
Latent space mapping (RAVE)                   Defined concepts
         │                                              │
         ▼                                              ▼
Oracle validation (consensus, PKI)            Primitive concepts
         │                                              │
         ▼                                              ▼
Energy trade / currency rules                 Postulates / theories
         │                                              │
         └───────────────────┬──────────────────────────┘
                             │
               [ Oracle as Controlled Hallucination ]
    Anchor: external signals (energy fingerprints) constrain generation  
    Invocation: imagination emerges within constraints  
    Anil Seth: perception is controlled hallucination guided by input

From sound reproducibility to semantic stability

At present, the sound engine simply connects the 7-dimensional fingerprint to the RAVE decoder inlets. This minimal design already yields strong reproducibility: each solar cell produces a stable sonic identity whenever measured. Looking forward, methods from cross-modal embedding research (such as Jha et al. 2025 on vec2vec) suggest how this mapping could be formalized. They propose constraints like cycle-consistency and distance preservation to ensure that embeddings remain semantically stable across modalities. For this project, such ideas are not yet implemented, but they provide useful guidance for future experiments on how photovoltaic features might align more systematically with audio latent space.

• 

  A simple sound map diagram of I-V data and RAVE latent space.

The hardware challenge

There is no suitable on-device solution yet. Running a RAVE model on mobile or wearable hardware remains challenging. Jetson-class devices are possible, but power consumption makes them weak candidates for a solar walkman. Current focus stays on data quality, mapping priors, and stable sonification on a laptop host.

Discussion

Where things are now

1. The oracle walkman works as a simple art sculpture that sonifies DSSC I–V curves in real time. The 7-feature fingerprint is stable across illumination changes after normalization.
2. The mapping is deliberately minimal, which makes evaluation of reproducibility straightforward. A controlled pipeline from sensing to sound is established in Max/MSP.
3. Perception and AI are treated as two sides of the same generative mechanism. The working definition of hallucination is generation that drifts beyond admissible evidence and priors. Brains predict and correct; hallucination is an extreme case of prediction mismatch. The oracle provides external anchors to keep generation within verifiable bounds while leaving room for creative variance.

What the theory is doing now

1. Stinson’s generic-mechanism view motivates treating DSSC–RAVE and human perception as different instantiations of a common generative architecture.
2. Feigl’s correspondence model motivates explicit bridges from observation to latent variables, so every design step is tied back to measurable traces.
3. These theoretical lenses are not goals in themselves. They function as design guidelines for dataset building, priors for mapping, and evaluation metrics for drift and variance.

Next steps

1. Build a small but clean training set of DSSC fingerprints with controlled illumination and temperature, then test monotonicity and local smoothness priors.
2. Prototype vec2vec-style constraints: simple cycle checks and distance preservation on a held-out set; log when sonic neighborhoods fail to match energy-curve neighborhoods.
3. Investigate lightweight inference targets and compression for future mobile use.
4. Explore whether traceable energy records can be registered as verifiable hashes derived from sonic fingerprints, then evaluate failure modes and anti-counterfeiting limits.

The design of the solar mini disc walkman is therefore not only a music project. It also opens a measured way to ask whether energy can be traceable and traded, and how a physical fingerprint might act as an oracle that links matter, perception, and computation without collapsing artistic variability.

References

1. Buckner, Cameron J. 2023. From Deep Learning to Rational Machines: What the History of Philosophy Can Teach Us about the Future of Artificial Intelligence. 1st ed. Oxford University PressNew York. https://doi.org/10.1093/oso/9780197653302.001.0001.
2. Stinson, Catherine. 2020. “From Implausible Artificial Neurons to Idealized Cognitive Models: Rebooting Philosophy of Artificial Intelligence.” Philosophy of Science 87 (4): 590–611. https://doi.org/10.1086/709730.
3. Jha, Rishi, Collin Zhang, Vitaly Shmatikov, and John X. Morris. 2025. “Harnessing the Universal Geometry of Embeddings.” arXiv:2505.12540. Preprint, arXiv, June 25. https://doi.org/10.48550/arXiv.2505.12540.
4. https://www.hackteria.org/wiki/A_RAVE_and_starvation_synth_based_generative_sonic_device_powered_by_dye_sensitized_solar_cell
5. https://github.com/shihweichieh2023/IVcurve_tester
6. https://github.com/rjha18/vec2vec
7. https://medium.com/@shihweichieh/generative-systems-and-extended-mind-as-transformation-similarity-models-connecting-two-cultures-0352458fe85c