Record Realtek chirp routing authority

This commit is contained in:
Meta Vens 2026-06-01 03:13:29 +01:00
commit 60fb77241e
4 changed files with 20 additions and 5 deletions

View file

@ -25,8 +25,12 @@ flowchart TD
## Invariants
- Scarlett speaker loopback is the timing authority when calibration chirplets
are playing.
- Starfire Realtek is the active chirp speaker and loopback-measurement path in
the current routing. The Windows system default output is now Scarlett for
monitoring, so default-output chirps are wrong unless they explicitly target
Realtek.
- Scarlett ASIO is the professional mic and monitor interface. Its outputs are
not the current active chirp-emission authority.
- Focusrite dialogue mics are the voice anchors.
- Camera mics are spatial/context witnesses.
- Loopback/program audio is timing evidence where available; it outranks

View file

@ -111,9 +111,13 @@ not become the synchronized program authority.
## Audio Field
The six-microphone path is separate from the bridge. Scarlett speaker loopback
is the current timing authority when active calibration is playing, but Scarlett
production capture belongs on ASIO rather than WASAPI shared mode.
The six-microphone path is separate from the bridge. Current local routing is:
Windows system default output is Scarlett for monitoring, while Starfire
Realtek is the active chirp-emission and loopback-measurement path. Any active
chirp runner that emits to the default output is therefore wrong unless it
explicitly selects Realtek. Scarlett production capture belongs on ASIO rather
than WASAPI shared mode, and Scarlett remains the professional mic/monitor
interface rather than the current chirp speaker authority.
`MimirBioacousticTimeline` owns the active runtime watermark described in
[[docs/bioacoustic-timeline-watermark|Bioacoustic Timeline Watermark]]: a
low-gain birdsong-like word language with 128 self-identifying word positions,

View file

@ -311,3 +311,4 @@
{"ts": "2026-06-01T03:04:44+01:00", "type": "decision", "status": "ok", "note": "Targeted bioacoustic probing is now residual-driven instead of generic sweep-driven. MimirTargetedBioacousticProbePlanner consumes per-source spectral residuals and optional mic pathology notes such as shotgun noise floor and low-heavy tilt, then emits gain-bounded probe bands. The first smoke scopes authority to the 192 kHz Scarlett shotgun/cardioid pair and selects about 18.6 kHz unmeasured high-band evidence first, then about 9.9 kHz weak response, 3.6 kHz phase instability, and 910 Hz magnitude residual. Wider room-field use of every connected mic remains later evidence, not the initial calibration owner."}
{"ts": "2026-06-01T03:09:15+01:00", "type": "verification", "status": "ok", "note": "Live targeted bioacoustic probe ran through Focusrite ASIO at 192 kHz with the shotgun/cardioid pair. Gain 1.0 artifact artifacts/asio/targeted-bioacoustic-live/20260601-030836 selected 18.6 kHz, 9.9 kHz, 3.6 kHz, and 910 Hz; shotgun measured low/high tilt 20.199 dB and cardioid 15.082 dB. Gain 2.0 artifact artifacts/asio/targeted-bioacoustic-live/20260601-030857 doubled loopback energy but physical 18.6 kHz remained nearly absent: shotgun -51.475 dB vs loopback, cardioid -56.585 dB. 910 Hz was the strongest physical band, especially shotgun at -21.994 dB. Treat current high-band path as weak/unusable until a better speaker/mic placement or probe shape proves otherwise."}
{"ts": "2026-06-01T03:11:14+01:00", "type": "verification", "status": "ok", "note": "Louder targeted bioacoustic probes ran at playback gains 4.0 and 8.0. Gain 4 artifact artifacts/asio/targeted-bioacoustic-live/20260601-031030 had shotgun peak 0.020941 and cardioid peak 0.016305; gain 8 artifact artifacts/asio/targeted-bioacoustic-live/20260601-031051 had shotgun peak 0.140715 and cardioid peak 0.027774, so the pass was audible but not clipped. Even at gain 8, 18.6 kHz stayed unusable: shotgun -54.910 dB and cardioid -63.258 dB vs loopback. Shotgun 910 Hz reached -26.549 dB with 0.237 confidence and 3.6 kHz reached -32.637 dB with 0.082 confidence. Current probe loop should bias usable low/mid anchors while treating ultrasonic/high band as failed path evidence."}
{"ts": "2026-06-01T03:13:10+01:00", "type": "decision", "status": "ok", "note": "Routing correction: Starfire system default output is now Scarlett for monitoring, while Starfire Realtek is the intended active chirp-emission and loopback-measurement path. The 20260601 targeted ASIO-output probe artifacts remain useful Scarlett-output diagnostics but are not the intended Realtek chirp path. Next active probe runner must explicitly select Realtek output/loopback and must not assume default output or Scarlett ASIO playback."}

View file

@ -673,6 +673,12 @@ accepted_design:
stayed unusable at shotgun -54.910 dB and cardioid -63.258 dB relative to
loopback. The current physical path should bias usable low/mid anchors
while treating ultrasonic/high-band probes as failed-path evidence.
Routing correction: Starfire's Windows system default output is now
Scarlett for monitoring, while Starfire Realtek is the intended
chirp-emission and loopback-measurement path. The 20260601 targeted
ASIO-output artifacts remain useful Scarlett-output diagnostics, but the
next active probe runner must explicitly target Realtek output/loopback
instead of assuming default output or Scarlett ASIO playback.
A fresh live Scarlett proof is preserved at
`calibration/bioacoustic/complex-contour-live-20260525-063229.md`:
Focusrite ASIO playback/capture measured Loopback 1 -> Loopback 2 at