feat(host/audio): the minted microphone returns to tier-0 — pitch-true
ci / web (pull_request) Successful in 1m30s
apple / swift (pull_request) Successful in 1m26s
apple / screenshots (pull_request) Skipped
ci / bun-nix (pull_request) Successful in 21s
ci / docs-site (pull_request) Successful in 2m18s
android / android (pull_request) Successful in 3m27s
ci / rust-arm64 (pull_request) Successful in 4m24s
ci / rust (pull_request) Canceled after 7m47s

The revert un-reverts, on measurement: with the per-direction stamp sets
(render = the pad-proven PCM16-device/float-mix stereo split, capture =
device-format only), micpitch reads 440 Hz in as 440 Hz out at exact
peak. The octave-low voice was the driver DEFAULT endpoints disagreeing
(stereo render vs mono capture), never a raw-crossing design. The user
called the wrong verdict — the pad program 4ch success was the
counter-evidence that reopened the case.
This commit is contained in:
2026-08-07 16:53:48 +02:00
parent 52cd42e911
commit ebc2f1cf92
@@ -109,20 +109,17 @@ static LAST_ATTEMPT: Mutex<Option<Instant>> = Mutex::new(None);
/// The wiring plan's tier-0 input: the minted ids, or all-empty while nothing is provisioned.
///
/// **The mic ids are deliberately NOT published** (2026-08-07 pitch measurements): the SSM
/// driver's pins are format-locked (render stereo-only, capture mono-only) and the crossing
/// is a RAW byte pass, so voice fed through the render endpoint reads back an octave low and
/// no stamp can fix it — S3's peak-based PASS was a false pass. The mic therefore falls back
/// to the name ladder (a virtual cable) per the design's revert clause, while the SPEAKERS
/// substrate — which involves no driver crossing, just an engine loopback tap — stays tier-0.
/// The mic endpoints are still minted and recorded ([`provisioned`]) for the `micpitch`
/// probe and for a future transport that bypasses the render path.
/// The mic ids were briefly unpublished during the 2026-08-07 pitch investigation ("voice an
/// octave low") — the eventual measured truth: both pins run stereo/48 kHz fine, the octave
/// came from the driver's DEFAULT endpoints disagreeing (stereo render vs mono capture), and
/// the per-direction stamp sets in [`stamp_identity`] fix it permanently
/// (`audio-probe micpitch`: 440 Hz in → 440 Hz out, peak exact). Full tier-0 restored.
pub(crate) fn minted_ids() -> wiring_plan::MintedIds {
match PROVISIONED.get() {
Some(m) => wiring_plan::MintedIds {
speakers_render: m.speakers_render.clone(),
mic_render: None,
mic_capture: None,
mic_render: m.mic_render.clone(),
mic_capture: m.mic_capture.clone(),
},
None => wiring_plan::MintedIds::default(),
}