ELUDGet early access

The engineering journal.

What we're working on, what we've learned and what still has to be proven. We'll post test results here, including the ones that don't go our way.

Now

Design and early prototypes

  • Industrial design and 3D-printed, hand-assembled ELUD One prototypes
  • Early-access interest registration is open
Next

Prove it works

  • Test effectiveness in everyday places, safety and battery life
  • Publish the method first, then the results
Then

Refundable reservations

  • Open to early-access members first
  • Deposit amount shown before anyone pays
Later

Small first batch

  • Small-batch production of ELUD One
  • Timing shared only once we can stand behind it

Notes

  1. October 24, 2026 · ELUD engineering

    A quieter design language

    We've settled on a soft, warm and monochrome look for everything ELUD makes. Privacy technology shouldn't announce itself, and neither should its brand.

    Brand
  2. October 23, 2026 · ELUD product development

    What the first ELUD One prototype needs to prove

    We are preparing the physical design and testing plan. Acoustic effectiveness, battery life, safety and build repeatability remain open questions; no performance results are available yet.

    Development

Methods / First, the baseline

How we will test ELUD One.

The question is not whether a visual waveform looks quieter. It is whether an independent microphone captures less intelligible speech under matched, documented conditions—and what tradeoffs the device introduces.

Protocol proposed October 24, 2026. No reference or activated-device recordings are published yet.

  1. Hold the conditions constant. Record the same speech source, microphone model, gain, placement, distance, room, ambient level and device position.
  2. Capture reference and active trials. Preserve original audio, matched timestamps and a test identifier; log device mode, battery state and output level.
  3. Measure what matters. Compare amplitude envelopes, speech intelligibility, audibility to people nearby and repeatability. Report failures as well as improvements.
  4. Check safety and practical limits. Evaluate listener exposure, output levels, battery endurance and behavior across room types before any product claim.
  5. Publish provenance with the results. Disclose microphone, distance, ambient levels, method, date, and enough recording metadata to interpret findings.

Follow the work as it happens.

Our development notes are available here and via RSS. Register for early access to hear about the next engineering milestone.