0
Fork 0
mirror of https://github.com/zeronsh/comet.git synced 2026-11-16 18:57:19 +00:00
comet/docs/sound-design
Gaël Cado 52cc9ab80e
Add a configurable Zeron sound signature (#320)
* Introduce rounded completion and question notification sounds

* Keep additional rounded sound auditions on notification branch

* Wire configurable notification sound signature

* Make notification switches keyboard accessible

* fix notification alert edge cases

---------

Co-authored-by: wing-anara <wing@anara.com>
History 2026-09-12 17:48:48 -07:00
..
auditions Add a configurable Zeron sound signature (#320) 2026-09-12 17:48:48 -07:00
README.md Add a configurable Zeron sound signature (#320) 2026-09-12 17:48:48 -07:00

Rounded notification sounds

This contribution selects a compact four-cue product signature. It replaces the existing completion and agent-question sounds, adds one shared attention cue for run failures and durable connection outages, and preserves the Appshot capture cue from the separate Appshots contribution.

  • done.wav: completion, settling rounded pair.
  • request.wav: agent question, rising rounded pair.
  • attention.wav: failure or durable disconnection, restrained downward rounded pair.
  • generate-notification-sounds.py: reproduces the completion and question assets; --variants-dir <directory> also exports the two click-only alternatives.

Settings → Notifications has a master session-sounds switch and independent choices for completion, input required, and errors/disconnections. Appshot capture keeps its independent Capture sound setting on the Appshots page.

Additional auditions

Ten additional variants cover send, queue, upload ready, completion, questions, attention, microphone on/off, reconnection and undo. Only the attention audition is promoted to an application asset; the rest remain references without triggers.

Earlier click-only alternatives: completion and agent question.

Run python3 scripts/generate-sound-auditions.py --install-attention to regenerate the ten numbered WAVs, their manifest, and the selected attention asset. All synthesis uses Python's standard library and original rounded pressure pulses, with no external samples.