diff --git a/cli/XAir.c b/cli/XAir.c
index 1ce2978..e929251 100644
--- a/cli/XAir.c
+++ b/cli/XAir.c
@@ -121,7 +121,7 @@ static void handle_fx_type(int slot) {
 }
 
 static void handle_fx_par(int slot) {
-    /* echo the set command back to caller (X32 behaviour) */
+    /* echo the set command back to caller (X32 behavior) */
     memcpy(s_buf, r_buf, r_len);
     s_len = r_len;
     xsend((struct sockaddr *)&client_ip, client_ip_len);
diff --git a/cli/ringout_logic.c b/cli/ringout_logic.c
index 2dfb5dd..c45e0e0 100644
--- a/cli/ringout_logic.c
+++ b/cli/ringout_logic.c
@@ -44,7 +44,7 @@ RingoutAction ringout_step(RingoutState *st, const float *bins, const float *bas
     }
 
     /* A notch placed last step takes priority: back the gain off and re-settle
-     * before analysing again, so the notch can take effect. */
+     * before analyzing again, so the notch can take effect. */
     if (st->pending_backoff) {
         st->pending_backoff = 0;
         st->cur_gain_db -= st->cfg.step_db;
@@ -57,7 +57,7 @@ RingoutAction ringout_step(RingoutState *st, const float *bins, const float *bas
         return a;
     }
 
-    /* Settle: ignore frame content until the system has stabilised. */
+    /* Settle: ignore frame content until the system has stabilized. */
     if (st->phase == RO_PHASE_SETTLE) {
         if (st->settle_count > 0) {
             st->settle_count--;
@@ -67,10 +67,10 @@ RingoutAction ringout_step(RingoutState *st, const float *bins, const float *bas
         st->confirm_count = 0;
         st->confirm_bin   = -1;
         st->stable_count  = 0;
-        /* fall through and analyse this frame */
+        /* fall through and analyze this frame */
     }
 
-    /* Analyse: is there a narrow ring? */
+    /* Analyze: is there a narrow ring? */
     int peak;
     int found  = detect_peak(bins, baseline, st->cfg.threshold_db, &peak);
     int narrow = found && is_narrow_peak(bins, RTA_BINS, peak,
diff --git a/cli/ringout_logic.h b/cli/ringout_logic.h
index a73b2c1..abe49ef 100644
--- a/cli/ringout_logic.h
+++ b/cli/ringout_logic.h
@@ -52,7 +52,7 @@ typedef struct {
     int   stable_frames;     /* consecutive ring-free frames required before raising */
     int   narrow_skip;       /* is_narrow_peak: skip */
     int   narrow_span;       /* is_narrow_peak: span */
-    float narrow_min_db;     /* is_narrow_peak: min advantage over neighbours */
+    float narrow_min_db;     /* is_narrow_peak: min advantage over neighbors */
 } RingoutConfig;
 
 typedef struct {
@@ -73,7 +73,7 @@ typedef struct {
     float margin_db;         /* gain-before-feedback margin achieved (cur - start) */
 } RingoutState;
 
-/* Initialise state from config. cfg is copied; sane fields are required (e.g.
+/* Initialize state from config. cfg is copied; sane fields are required (e.g.
  * step_db > 0, ceiling_db >= start_gain_db). The bus is assumed to already be at
  * cfg.start_gain_db when the loop begins. */
 void ringout_init(RingoutState *st, const RingoutConfig *cfg);
diff --git a/cli/rta_bins.c b/cli/rta_bins.c
index 8fa2c73..bf55d58 100644
--- a/cli/rta_bins.c
+++ b/cli/rta_bins.c
@@ -3,7 +3,7 @@
 
 /* Log-frequency interpolation of the 31 ISO 1/3-octave GEQ anchors
  * (TOAST_GEQ_BIN ↔ ISO centers), 1 kHz pinned to bin 56. Off-anchor bins are
- * modelled, not measured — verify with a console sweep via --rta-probe. */
+ * modeled, not measured — verify with a console sweep via --rta-probe. */
 const float RTA_BIN_FREQ[RTA_BIN_COUNT] = {
         20.0f,     21.5f,     23.2f,     25.0f,     26.5f,     28.1f,     29.7f,     31.5f,
         34.1f,     36.9f,     40.0f,     43.1f,     46.4f,     50.0f,     54.0f,     58.3f,
diff --git a/cli/rta_probe.py b/cli/rta_probe.py
index c1c04b6..16088bc 100644
--- a/cli/rta_probe.py
+++ b/cli/rta_probe.py
@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 """Live /meters/4 RTA peak-bin probe for X32/XAir — field tool for T-027.
 
-Subscribe to the 100-bin RTA, print the peak bin (and its modelled frequency
+Subscribe to the 100-bin RTA, print the peak bin (and its modeled frequency
 from cli/rta_bins.c). Use with a known tone to verify the bin->Hz table:
 a 1 kHz pistonphone must light bin 56; a console sine sweep verifies the rest.
 
@@ -15,7 +15,7 @@
     sys.exit("usage: rta_probe.py <xr18_ip> [seconds]")
 PORT = 10024
 
-# modelled bin->Hz, parsed from the C table so the two never drift
+# modeled bin->Hz, parsed from the C table so the two never drift
 try:
     txt = pathlib.Path(__file__).with_name("rta_bins.c").read_text()
     BIN_FREQ = [float(x) for x in re.findall(r"([0-9]+\.[0-9]+)f", txt)[:100]]
diff --git a/cli/test_ringout_emu.py b/cli/test_ringout_emu.py
index bed84ce..90a0999 100644
--- a/cli/test_ringout_emu.py
+++ b/cli/test_ringout_emu.py
@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 """T-025 / T-032 ring-out safety harness: a mock X32 that models the feedback
 loop and asserts the ring-out control loop's safety properties (ceiling, notch,
-restore, abort). No audio — feedback is modelled as a function of commanded
+restore, abort). No audio — feedback is modeled as a function of commanded
 bus gain via a pluggable RoomModel.
 
 Run:
diff --git a/docs/ringout-design.md b/docs/ringout-design.md
index 7cd5546..d7cdc28 100644
--- a/docs/ringout-design.md
+++ b/docs/ringout-design.md
@@ -16,7 +16,7 @@ with a handoff:
   **static notch profile + the measured gain-before-feedback margin**.
 - **Reactive (protect)** — starts *from that profile* (static notches pre-placed)
   and dynamically defends the remaining margin during the show. This is the
-  existing ToastSaver behaviour, plus a new "load a profile at startup" step.
+  existing ToastSaver behavior, plus a new "load a profile at startup" step.
 
 > The proactive pass *produces* what the reactive pass *consumes*. "React as
 > planned" = the planned notches are already in when the show starts; live mode
diff --git a/docs/system-tune-design.md b/docs/system-tune-design.md
index 56fd61e..416ff03 100644
--- a/docs/system-tune-design.md
+++ b/docs/system-tune-design.md
@@ -55,7 +55,7 @@ generator and writes the GEQ.
 1. **Mains response EQ** — pink noise → measure at a mic → flatten LR GEQ to the
    target curve.
 2. **Per-monitor** — optionally pink-flatten each wedge, **then ring it out**
-   (the current work) to maximise gain-before-feedback.
+   (the current work) to maximize gain-before-feedback.
 3. **Report** — final EQ curves + per-monitor headroom margins (and reuse the
    ring-out profile format from `T-024`).
 
diff --git a/esp32/X32Link/beat_synth.h b/esp32/X32Link/beat_synth.h
index 78244fd..7c4ee00 100644
--- a/esp32/X32Link/beat_synth.h
+++ b/esp32/X32Link/beat_synth.h
@@ -16,14 +16,14 @@ typedef struct {
     uint32_t last_ms;   // wall-clock ms of the last emitted beat
 } BeatSynth;
 
-// Clear state (last_ms = 0). Behaviour then matches a fresh static counter:
+// Clear state (last_ms = 0). Behavior then matches a fresh static counter:
 // the first step() with now_ms past one interval fires immediately.
 void beat_synth_reset(BeatSynth* s);
 
 // Advance to now_ms and return true if a beat is due at `bpm`. Fires when
 // (now_ms - last_ms) has reached the 60000/bpm interval (unsigned/wrap-safe),
 // then latches last_ms = now_ms. bpm <= 0 always returns false and leaves state
-// untouched (so a resumed BPM fires promptly, matching the old inline behaviour).
+// untouched (so a resumed BPM fires promptly, matching the old inline behavior).
 bool beat_synth_step(BeatSynth* s, uint32_t now_ms, float bpm);
 
 #ifdef __cplusplus
diff --git a/esp32/X32Link/web_config.cpp b/esp32/X32Link/web_config.cpp
index 8dd1376..2baab96 100644
--- a/esp32/X32Link/web_config.cpp
+++ b/esp32/X32Link/web_config.cpp
@@ -178,7 +178,7 @@ if(curSlot>maxSlot){curSlot=maxSlot;hSlot.value=maxSlot}renderSlots()});
 renderSlots();
 var bpmEl=document.getElementById('bpm'),beatEl=document.getElementById('beat');
 var seedBpm=parseFloat(bpmEl.textContent)||0,beatTimer=null,shownBpm=-1;
-// Free-running fallback blink (pre-LNK-022 behaviour) — used whenever the
+// Free-running fallback blink (pre-LNK-022 behavior) — used whenever the
 // server says phase isn't valid yet (sync gap / no source). Unchanged.
 function setBeat(bpm){if(beatTimer){clearInterval(beatTimer);beatTimer=null}
 if(bpm>0){beatTimer=setInterval(function(){beatEl.classList.add('on');setTimeout(function(){beatEl.classList.remove('on')},90)},60000/bpm)}}
diff --git a/esp32/X32MidiClock/X32MidiClock.ino b/esp32/X32MidiClock/X32MidiClock.ino
index bddd53b..e3c8e20 100644
--- a/esp32/X32MidiClock/X32MidiClock.ino
+++ b/esp32/X32MidiClock/X32MidiClock.ino
@@ -112,7 +112,7 @@ void setup() {
                    g_config.fx_slot, g_config.mixer_ip,
                    config_model_port(g_config.model));
 
-    // USB MIDI must be initialised before WiFi to ensure USB enumeration is fast
+    // USB MIDI must be initialized before WiFi to ensure USB enumeration is fast
     // Device name is set in midi_clock.cpp via USBMIDI constructor
     USB.manufacturerName("X32Link");
     midi_clock_init();  // calls MidiUSB.begin() + spawns poll task
diff --git a/esp32/X32MidiClock/midi_bpm.cpp b/esp32/X32MidiClock/midi_bpm.cpp
index e05c9c6..370d1dc 100644
--- a/esp32/X32MidiClock/midi_bpm.cpp
+++ b/esp32/X32MidiClock/midi_bpm.cpp
@@ -5,7 +5,7 @@
 #include <Arduino.h>
 
 void midi_bpm_init() {
-    // nothing to initialise — state lives in midi_clock module
+    // nothing to initialize — state lives in midi_clock module
 }
 
 float midi_bpm_update() {
diff --git a/esp32/tasks/FDR-002.md b/esp32/tasks/FDR-002.md
index fd5122f..eaec911 100644
--- a/esp32/tasks/FDR-002.md
+++ b/esp32/tasks/FDR-002.md
@@ -16,7 +16,7 @@ updated_at: 2026-06-29
 
 ## Goal
 
-Implement the FaderDisp behaviour: when a channel fader moves, show its value in
+Implement the FaderDisp behavior: when a channel fader moves, show its value in
 **dB on that channel's scribble strip**, by writing `/ch/NN/config/name`. On a
 clean shutdown / disable, restore the original names.
 
diff --git a/esp32/tasks/ITT-002.md b/esp32/tasks/ITT-002.md
index 57a77a0..704fc18 100644
--- a/esp32/tasks/ITT-002.md
+++ b/esp32/tasks/ITT-002.md
@@ -45,5 +45,5 @@ send a MIDI message). This is the "If This Then That" core.
   and send as the raw OSC float (the mixer applies its own fader taper).
 - Keep the engine allocation-free at eval time; build the rule table once at
   load / on web save.
-- This generalises X32Midi2OSC (MIDI→OSC) and X32Commander (OSC→MIDI/OSC) from
+- This generalizes X32Midi2OSC (MIDI→OSC) and X32Commander (OSC→MIDI/OSC) from
   the upstream toolset into one bidirectional firmware.
diff --git a/esp32/tasks/LNK-001.md b/esp32/tasks/LNK-001.md
index 06022b4..520ed5f 100644
--- a/esp32/tasks/LNK-001.md
+++ b/esp32/tasks/LNK-001.md
@@ -13,7 +13,7 @@ updated_at: 2026-06-26
 ## Goal
 
 Create the `esp32/X32Link/` Arduino sketch directory with a minimal WiFi
-sketch and a `config.h` that centralises all runtime parameters. Verify the
+sketch and a `config.h` that centralizes all runtime parameters. Verify the
 bare sketch compiles and connects to WiFi before any Link or OSC code is added.
 
 This task targets any ESP32-S3 board (XIAO ESP32-S3, ESP32-S3 Super Mini, etc.).
@@ -71,7 +71,7 @@ arduino-cli upload \
 
 **Serial quirk**: On ESP32-S3 with USB CDC, open the serial port with
 DTR=True, RTS=False BEFORE pressing RST. Hardware reset required after upload
-(same behaviour as ESP-001 / XIAO ESP32-S3).
+(same behavior as ESP-001 / XIAO ESP32-S3).
 
 **config.h constants** — all values as `#define`:
 
diff --git a/esp32/tasks/LNK-003.md b/esp32/tasks/LNK-003.md
index ea0b04d..7e0a80d 100644
--- a/esp32/tasks/LNK-003.md
+++ b/esp32/tasks/LNK-003.md
@@ -1,6 +1,6 @@
 ---
 id: LNK-003
-title: WiFi connection and Link session initialisation
+title: WiFi connection and Link session initialization
 status: done
 assignee: null
 priority: high
diff --git a/esp32/tasks/LNK-005.md b/esp32/tasks/LNK-005.md
index 3695b6d..5833753 100644
--- a/esp32/tasks/LNK-005.md
+++ b/esp32/tasks/LNK-005.md
@@ -12,7 +12,7 @@ updated_at: 2026-06-26
 
 ## Goal
 
-Open a UDP socket to the XR18 and write the current BPM as a normalised float
+Open a UDP socket to the XR18 and write the current BPM as a normalized float
 to `/fx/{slot}/par/02` whenever `g_bpm_changed` is set. This is the final
 output stage — Link tempo → XR18 delay time.
 
@@ -23,13 +23,13 @@ output stage — Link tempo → XR18 delay time.
 - [ ] `send_bpm_osc()` builds a valid OSC `/fx/{slot}/par/02 ,f {val}` packet
 - [ ] OSC packet byte layout verified against `src/X32lib/Xsprint.c` reference
 - [ ] XR18 FX slot delay time visibly changes in X-Air Edit when DAW BPM changes
-- [ ] BPM range tested: 60, 120, 180 BPM → correct normalised values (see Notes)
+- [ ] BPM range tested: 60, 120, 180 BPM → correct normalized values (see Notes)
 
 ## Notes
 
 **Firmware scope**: `esp32/X32Link/` only. Do NOT touch `esp32/X32_emulator/`.
 
-**BPM → normalised float** (inline — cannot share `src/x32tap_logic.c`):
+**BPM → normalized float** (inline — cannot share `src/x32tap_logic.c`):
 ```c
 // Mirrors src/x32tap_logic.c: ms_to_normalized(), TAP_MAX_MS = 3000
 static float bpm_to_normalized(float bpm) {
@@ -41,7 +41,7 @@ static float bpm_to_normalized(float bpm) {
 ```
 
 **Expected values**:
-| BPM  | delay_ms | normalised |
+| BPM  | delay_ms | normalized |
 |------|----------|------------|
 |  60  | 1000 ms  | 0.3333     |
 | 120  |  500 ms  | 0.1667     |
diff --git a/esp32/tasks/LNK-009.md b/esp32/tasks/LNK-009.md
index 6389502..79d9344 100644
--- a/esp32/tasks/LNK-009.md
+++ b/esp32/tasks/LNK-009.md
@@ -53,7 +53,7 @@ same knobs under different prefixes:
 
 ## Constraints
 
-- No behaviour change: same values reach the same call sites, just sourced
+- No behavior change: same values reach the same call sites, just sourced
   per-input instead of via a fixed macro.
 - `DEFAULT_WIFI_*` / `DEFAULT_MIXER_IP` / `DEFAULT_MODEL` / `DEFAULT_FX_SLOT`
   already share names — leave them.
diff --git a/esp32/tasks/LNK-011.md b/esp32/tasks/LNK-011.md
index 9a55ccc..f9cea73 100644
--- a/esp32/tasks/LNK-011.md
+++ b/esp32/tasks/LNK-011.md
@@ -60,7 +60,7 @@ PublishDecision bpm_publisher_step(float bpm, bool active, uint32_t now_ms);
 
 - Pure C, no Arduino deps — buildable and testable on host (follow the
   existing `test/test_link_protocol.c` + weak-`millis` pattern).
-- Preserve current behaviour exactly: bpm=0 ⇒ no send (mixer holds last value);
+- Preserve current behavior exactly: bpm=0 ⇒ no send (mixer holds last value);
   send only on change-past-threshold OR per-bar refresh; respect min interval.
 
 ## Acceptance Criteria
diff --git a/esp32/tasks/LNK-012.md b/esp32/tasks/LNK-012.md
index 3e18e22..5718e59 100644
--- a/esp32/tasks/LNK-012.md
+++ b/esp32/tasks/LNK-012.md
@@ -34,7 +34,7 @@ no input-source field yet.
    `TempoSourceKind` enum from LNK-008. Default = Link.
 2. **NVS** (`app_config_nvs.cpp`): load/save the new key `input_src`; default to
    Link when absent (first boot / post factory-reset).
-3. **Web form** (`web_config.cpp`): add a labelled dropdown
+3. **Web form** (`web_config.cpp`): add a labeled dropdown
    "Tempo source: Ableton Link / USB MIDI clock"; submit persists it like the
    other fields, then reboots (existing save→restart flow).
 4. **Boot** (LNK-010 `setup`): `tempo_source_select(cfg.input_source)` before
diff --git a/esp32/tasks/LNK-031.md b/esp32/tasks/LNK-031.md
index 566fb14..cc2d582 100644
--- a/esp32/tasks/LNK-031.md
+++ b/esp32/tasks/LNK-031.md
@@ -61,7 +61,7 @@ in `link_measurement.c` stay as-is and are called by the session or the glue.
       timeouts emit END_FAIL; different-peer and due-timer both retrigger; trigger
       gated by `active`; pong carries prev-ghost / commits at READY.
 - [x] `link_measurement_io.cpp` reduced to sockets + timing that translate the
-      session's actions (`run_action`); behaviour faithfully ported (LNK-026
+      session's actions (`run_action`); behavior faithfully ported (LNK-026
       fixes preserved as session actions).
 - [x] Full host suite green (20 suites); firmware compiles 86%; `AGENTS.md` updated.
 - [x] **Re-verified on hardware** (flashed; Link peer @ 127 BPM). Steady play:
diff --git a/esp32/tasks/MCK-005.md b/esp32/tasks/MCK-005.md
index 735afcc..91c6ec3 100644
--- a/esp32/tasks/MCK-005.md
+++ b/esp32/tasks/MCK-005.md
@@ -56,7 +56,7 @@ buffer just stays empty when disconnected; `midi_bpm_update()` returns 0.0 after
 the timeout, and OSC stops. On reconnect, Ableton re-sends clock and the
 firmware picks it back up within one beat.
 
-**Clock-stop behaviour (design decision)**: When the musician stops the DAW,
+**Clock-stop behavior (design decision)**: When the musician stops the DAW,
 we intentionally do NOT reset the X32 tempo to zero or a default. The desk
 keeps the last known tempo — the band can keep playing in time if the DAW
 stops unexpectedly. Tempo only changes when a new valid BPM arrives.
diff --git a/esp32/tasks/MCK-006.md b/esp32/tasks/MCK-006.md
index f5d487f..2e67807 100644
--- a/esp32/tasks/MCK-006.md
+++ b/esp32/tasks/MCK-006.md
@@ -24,7 +24,7 @@ DAW start/stop.
 - [ ] Enabling "Clock" output in Ableton and pressing Play causes FX delay time
       on XR18 to update within 1–2 beats
 - [ ] BPM accuracy: set Ableton to 80, 120, 140, 180 BPM → X-Air Edit delay time
-      matches expected normalised values (see LNK-005 reference table)
+      matches expected normalized values (see LNK-005 reference table)
 - [ ] Tempo change: drag Ableton BPM from 120 → 140 → X32 follows within 1 beat
 - [ ] DAW stop: FX delay time stays at last known value (does not reset)
 - [ ] DAW start after stop: tempo resumes correctly within 1 beat
diff --git a/esp32/test/test_beat_synth.c b/esp32/test/test_beat_synth.c
index b011033..1461970 100644
--- a/esp32/test/test_beat_synth.c
+++ b/esp32/test/test_beat_synth.c
@@ -7,7 +7,7 @@ void setUp(void)    { beat_synth_reset(&s); }
 void tearDown(void) {}
 
 void test_fresh_state_fires_on_first_step_past_interval(void) {
-    // Matches the old inline behaviour: static last=0, a large millis() reading
+    // Matches the old inline behavior: static last=0, a large millis() reading
     // is >= one interval so the first observed beat fires.
     TEST_ASSERT_TRUE(beat_synth_step(&s, 1000, 120.0f));  // interval 500, 1000-0 >= 500
 }
diff --git a/ios/ToastSaverCore/Sources/ToastSaverCore/NarrownessFilter.swift b/ios/ToastSaverCore/Sources/ToastSaverCore/NarrownessFilter.swift
index 96073a4..6ddd03d 100644
--- a/ios/ToastSaverCore/Sources/ToastSaverCore/NarrownessFilter.swift
+++ b/ios/ToastSaverCore/Sources/ToastSaverCore/NarrownessFilter.swift
@@ -1,11 +1,11 @@
 // Distinguishes a narrow feedback ring from broadband content (e.g. a vocal
-// formant): a real ring towers over its neighbours a few bins out. Ported from
+// formant): a real ring towers over its neighbors a few bins out. Ported from
 // cli/toast_logic.c (is_narrow_peak).
 
 public enum NarrownessFilter {
     /// True if `bins[peakBin]` is at least `minDB` above the average of the
-    /// neighbour bins sampled `skip+1 … skip+span` positions away on each side
-    /// (clamped to the array). Needs ≥1 neighbour on each side, else false.
+    /// neighbor bins sampled `skip+1 … skip+span` positions away on each side
+    /// (clamped to the array). Needs ≥1 neighbor on each side, else false.
     public static func isNarrowPeak(bins: [Float], peakBin: Int,
                                     skip: Int = 2, span: Int = 3, minDB: Float = 10.0) -> Bool {
         let n = bins.count
diff --git a/ios/ToastSaverCore/Sources/ToastSaverCore/RingoutController.swift b/ios/ToastSaverCore/Sources/ToastSaverCore/RingoutController.swift
index 6ba0d20..7e1069f 100644
--- a/ios/ToastSaverCore/Sources/ToastSaverCore/RingoutController.swift
+++ b/ios/ToastSaverCore/Sources/ToastSaverCore/RingoutController.swift
@@ -21,7 +21,7 @@ public struct RingoutConfig {
     public var cutDB: Float
     /// Consecutive frames a ring must persist before a notch fires.
     public var confirmFrames: Int
-    /// Frames to wait at current gain after any raise or notch before analysing.
+    /// Frames to wait at current gain after any raise or notch before analyzing.
     public var settleFrames: Int
     /// Consecutive ring-free frames required before raising gain.
     public var stableFrames: Int
@@ -29,7 +29,7 @@ public struct RingoutConfig {
     public var narrowSkip: Int
     /// Narrowness gate: bins to sample beyond the skip zone on each side.
     public var narrowSpan: Int
-    /// Narrowness gate: peak must exceed neighbour average by at least this many dB.
+    /// Narrowness gate: peak must exceed neighbor average by at least this many dB.
     public var narrowMinDB: Float
 
     public init(
@@ -147,17 +147,17 @@ public final class RingoutController {
             return .backOff(to: currentGainDB)
         }
 
-        // Settling: hold until the system stabilises after a raise or notch.
+        // Settling: hold until the system stabilizes after a raise or notch.
         if phase == .settle {
             if settleCount > 0 { settleCount -= 1; return .hold }
-            // Settle complete — fall through and analyse this frame.
+            // Settle complete — fall through and analyze this frame.
             phase        = .analyze
             confirmCount = 0
             confirmBin   = -1
             stableCount  = 0
         }
 
-        // Analyse: look for a narrow ring.
+        // Analyze: look for a narrow ring.
         guard let peakIndex = PeakDetector.detectPeak(
             bins: geqLevels, baseline: baseline, threshold: config.thresholdDB)
         else {
diff --git a/ios/ToastSaverCore/Tests/ToastSaverCoreTests/FeedbackDetectorTests.swift b/ios/ToastSaverCore/Tests/ToastSaverCoreTests/FeedbackDetectorTests.swift
index d99c679..086653f 100644
--- a/ios/ToastSaverCore/Tests/ToastSaverCoreTests/FeedbackDetectorTests.swift
+++ b/ios/ToastSaverCore/Tests/ToastSaverCoreTests/FeedbackDetectorTests.swift
@@ -38,7 +38,7 @@ final class FeedbackDetectorTests: XCTestCase {
         var b = flat(); b[idx] = spike; return b
     }
 
-    // narrowSkip=1, narrowSpan=1: a spike with at least 1 clear neighbour each side qualifies.
+    // narrowSkip=1, narrowSpan=1: a spike with at least 1 clear neighbor each side qualifies.
     private func makeDetector(confirm: Int = 3) -> FeedbackDetector {
         FeedbackDetector(threshold: 20.0, confirmFrames: confirm,
                          narrowSkip: 1, narrowSpan: 1, narrowMinDB: 5.0)
diff --git a/ios/ToastSaverCore/Tests/ToastSaverCoreTests/RingoutControllerTests.swift b/ios/ToastSaverCore/Tests/ToastSaverCoreTests/RingoutControllerTests.swift
index aeccd01..266cc6f 100644
--- a/ios/ToastSaverCore/Tests/ToastSaverCoreTests/RingoutControllerTests.swift
+++ b/ios/ToastSaverCore/Tests/ToastSaverCoreTests/RingoutControllerTests.swift
@@ -99,7 +99,7 @@ final class RingoutControllerTests: XCTestCase {
         _ = ctrl.step(geqLevels: quiet, baseline: baseline)  // raise
 
         // Inject a narrow spike on par 10 (index 9).
-        // With narrowSkip=1, narrowSpan=1: neighbours are at ±2, which exist.
+        // With narrowSkip=1, narrowSpan=1: neighbors are at ±2, which exist.
         let spike = spikedLevels(at: 9, spike: -10.0, floor: -80.0)
 
         var action: RingoutAction = .hold
diff --git a/ios/tasks/IOS-003.md b/ios/tasks/IOS-003.md
index b55a4ec..0e77ea0 100644
--- a/ios/tasks/IOS-003.md
+++ b/ios/tasks/IOS-003.md
@@ -35,12 +35,12 @@ final class FFTEngine {
     func process(_ samples: [Float]) -> [Float]
 
     /// Map an FFT bin index (0 ..< fftSize/2) to the nearest GEQ par (1–31).
-    /// Uses the same 1/3-octave centre frequencies as GEQMap.bins,
+    /// Uses the same 1/3-octave center frequencies as GEQMap.bins,
     /// recalculated for this engine's fftSize and sampleRate.
     func geqPar(forBin bin: Int) -> Int
 
     /// Precomputed GEQ bin indices for fftSize and sampleRate.
-    /// geqBins[i] is the FFT bin index closest to GEQ par i+1's centre frequency.
+    /// geqBins[i] is the FFT bin index closest to GEQ par i+1's center frequency.
     private(set) var geqBins: [Int]  // length 31
 }
 ```
@@ -87,7 +87,7 @@ Create `fftSetup` with `vDSP_create_fftsetup(log2n, FFTRadix(kFFTRadix2))` in `i
 
 ## geqBins calculation
 
-GEQ centre frequencies in Hz (31 bands, 1/3-octave, ISO):
+GEQ center frequencies in Hz (31 bands, 1/3-octave, ISO):
 ```
 [20, 25, 31.5, 40, 50, 63, 80, 100, 125, 160, 200, 250, 315, 400, 500,
  630, 800, 1000, 1250, 1600, 2000, 2500, 3150, 4000, 5000, 6300, 8000,
@@ -115,7 +115,7 @@ private func computeGeqBins() -> [Int] {
 
 ## Acceptance Criteria
 
-- [ ] `FFTEngine(fftSize: 2048, sampleRate: 48000)` initialises without crashing.
+- [ ] `FFTEngine(fftSize: 2048, sampleRate: 48000)` initializes without crashing.
 - [ ] `process(_:)` with a 1 kHz sine wave at 48 kHz, 2048 samples: peak bin is within ±2 bins of `Int(round(1000 * 2048 / 48000))` = bin 43.
 - [ ] `geqBins[17]` (1 kHz, 0-based index) ≈ 43 for 2048/48000.
 - [ ] `process(_:)` with all-zero input returns all values ≤ -80 dBFS.
diff --git a/ios/tasks/IOS-006.md b/ios/tasks/IOS-006.md
index 794de0c..ddea900 100644
--- a/ios/tasks/IOS-006.md
+++ b/ios/tasks/IOS-006.md
@@ -147,7 +147,7 @@ final class DetectorViewModel: ObservableObject {
 }
 ```
 
-`BaselineTracker` state (a `[Float]` array initialised to `Array(repeating: -60, count: spectrum.count)`) lives in `DetectorViewModel`; it is updated each frame before calling `detector.process`.
+`BaselineTracker` state (a `[Float]` array initialized to `Array(repeating: -60, count: spectrum.count)`) lives in `DetectorViewModel`; it is updated each frame before calling `detector.process`.
 
 ---
 
diff --git a/ios/tasks/IOS-008.md b/ios/tasks/IOS-008.md
index 79ceb9b..902b760 100644
--- a/ios/tasks/IOS-008.md
+++ b/ios/tasks/IOS-008.md
@@ -69,7 +69,7 @@ public struct RingoutConfig {
     /// Consecutive frames a peak must persist before a notch fires.
     public var confirmFrames: Int
 
-    /// Frames to wait (at current gain) after a raise before analysing again.
+    /// Frames to wait (at current gain) after a raise before analyzing again.
     public var settleFrames: Int
 
     /// Consecutive quiet frames required before declaring the gain level stable.
@@ -81,7 +81,7 @@ public struct RingoutConfig {
     /// Narrowness gate: sample this many bins beyond the skip zone on each side.
     public var narrowSpan: Int
 
-    /// Narrowness gate: peak must exceed neighbour average by at least this many dB.
+    /// Narrowness gate: peak must exceed neighbor average by at least this many dB.
     public var narrowMinDB: Float
 
     public init(
@@ -161,10 +161,10 @@ public final class RingoutController {
 
 Port the `ringout_step()` function from `cli/ringout_logic.c` faithfully. Key states:
 
-1. **Settling** — wait `settleFrames` frames after each gain raise before analysing.
+1. **Settling** — wait `settleFrames` frames after each gain raise before analyzing.
    Return `.hold` every frame. Decrement counter.
 
-2. **Analysing** — look for a narrow peak above threshold:
+2. **Analyzing** — look for a narrow peak above threshold:
    - Use `FeedbackDetector.detectPeak(bins:baseline:threshold:)` on the 31-element
      `geqLevels` array (it works on any `[Float]`, the 31 vs 100 length doesn't matter).
    - If a peak is found, pass through `NarrownessFilter.isNarrowPeak` on `geqLevels`.
diff --git a/ios/tasks/IOS-009.md b/ios/tasks/IOS-009.md
index e96b8e9..266723b 100644
--- a/ios/tasks/IOS-009.md
+++ b/ios/tasks/IOS-009.md
@@ -14,7 +14,7 @@ updated_at: 2026-06-29
 
 Implement `MixerConnection.swift` in the iOS app target. This supersedes the
 `XR18Connection` design from IOS-005 — do not build `XR18Connection`; build
-`MixerConnection` instead. It wraps `NWConnection` (UDP) and is parameterised by
+`MixerConnection` instead. It wraps `NWConnection` (UDP) and is parameterized by
 `MixerModel` so it works with both the XR18 (port 10024) and X32 (port 10023)
 without branching in the caller.
 
@@ -65,7 +65,7 @@ final class MixerConnection: ObservableObject {
     // MARK: - TEQ control
 
     /// Send /fx/{slot}/par/{par:02d} ,f {geqFloat(dB)}.
-    /// Fire-and-forget; does not wait for acknowledgement.
+    /// Fire-and-forget; does not wait for acknowledgment.
     func sendNotch(slot: Int, par: Int, dB: Float)
 
     /// Restore a TEQ band to flat (0 dB → geqFloat = 0.5).
@@ -137,7 +137,7 @@ func connect() async {
 
 Send `/xremote` every 9 seconds while connected. This matches the CLI's
 `XREMOTE_TIMEOUT` constant and keeps the mixer from unsubscribing the client.
-Both XR18 and X32 honour `/xremote`.
+Both XR18 and X32 honor `/xremote`.
 
 ### Fader read (`readFader`)
 
diff --git a/ios/tasks/IOS-010.md b/ios/tasks/IOS-010.md
index 26d65bf..59b6af5 100644
--- a/ios/tasks/IOS-010.md
+++ b/ios/tasks/IOS-010.md
@@ -206,7 +206,7 @@ private func runLoop(config: Config, initialBaseline: [Float]) async {
 
 ### Revert notches on abort
 
-Only revert if the run was aborted or interrupted — match CLI behaviour where
+Only revert if the run was aborted or interrupted — match CLI behavior where
 notches are left in place on clean completion:
 
 ```swift
diff --git a/tasks/T-021.md b/tasks/T-021.md
index 878e10d..d0af904 100644
--- a/tasks/T-021.md
+++ b/tasks/T-021.md
@@ -54,7 +54,7 @@ RingoutAction ringout_step(RingoutState *st, const float *bins, const float *bas
 - [x] `cli/ringout_logic.h` + `.c` implement `ringout_init` and `ringout_step`,
       reusing `detect_peak` / `is_narrow_peak` from `toast_logic` for ring
       detection (do **not** re-implement detection).
-- [x] State machine: after a gain change, wait `settle_frames`; analyse; if a
+- [x] State machine: after a gain change, wait `settle_frames`; analyze; if a
       confirmed narrow peak (`confirm_frames`) → `RO_PLACE_NOTCH` on its band +
       `RO_BACK_OFF`; else if stable → `RO_RAISE_GAIN` by `step_db`.
 - [x] **Safety in the pure layer:** never emit `RO_RAISE_GAIN` that would exceed
diff --git a/tasks/T-022.md b/tasks/T-022.md
index 92b48dd..381ef88 100644
--- a/tasks/T-022.md
+++ b/tasks/T-022.md
@@ -49,7 +49,7 @@ Fader float↔dB conversion: use the piecewise formula in
 - [x] RTA-tap guard: logs the configured RTA source channel and a reminder to
       ensure it taps the target monitor mic before ramping (best-effort warning;
       auto-select deferred).
-- [x] Reuses existing UDP send + `send_teq_band`; no change to reactive behaviour
+- [x] Reuses existing UDP send + `send_teq_band`; no change to reactive behavior
       (ring-out is a separate `run_ringout` flow gated on `--ringout`).
 
 ## Progress
diff --git a/tasks/T-023.md b/tasks/T-023.md
index d289383..3f31cf2 100644
--- a/tasks/T-023.md
+++ b/tasks/T-023.md
@@ -28,7 +28,7 @@ See [`docs/ringout-design.md`](../docs/ringout-design.md).
       `RO_PLACE_NOTCH` → `send_teq_band(par, db_to_geq_float(cut_db))` and record
       in `notches[]`; `RO_HOLD` → wait; `RO_DONE`/`RO_ABORT` → finish.
 - [x] **Supervised:** `--ringout-supervised` pauses for Enter before each
-      `RO_RAISE_GAIN` (`q` aborts); without it runs auto but still honours Ctrl-C.
+      `RO_RAISE_GAIN` (`q` aborts); without it runs auto but still honors Ctrl-C.
 - [x] Live progress to the terminal: per-action log lines with current driven
       gain, margin to ceiling, and notch placements.
 - [x] On completion print a summary: list of notched bands, **gain-before-feedback
diff --git a/tasks/T-024.md b/tasks/T-024.md
index 831b8e1..0b52ccf 100644
--- a/tasks/T-024.md
+++ b/tasks/T-024.md
@@ -26,7 +26,7 @@ See [`docs/ringout-design.md`](../docs/ringout-design.md) ("One pipeline").
       documented in `ringout_profile.c`.
 - [x] `--load-profile PATH` (reactive mode): parses the file, applies the saved
       notches via `send_teq_band`, then runs normal reactive detection on top.
-- [x] Loaded static notches carry a `from_profile` flag; **v1 behaviour: they are
+- [x] Loaded static notches carry a `from_profile` flag; **v1 behavior: they are
       pinned** (the release pass skips them) so the soundcheck notches stay put
       while live-detected notches still auto-release.
 - [x] Round-trip test: serialize → parse → re-serialize is byte-identical
diff --git a/tasks/T-025.md b/tasks/T-025.md
index afa64fc..693b0b4 100644
--- a/tasks/T-025.md
+++ b/tasks/T-025.md
@@ -24,7 +24,7 @@ See [`docs/ringout-design.md`](../docs/ringout-design.md) ("Safety").
       **synthetic feedback source**: either inject crafted `/meters/4` frames
       (a narrowband peak whose level tracks the commanded bus gain), or a small
       script that raises a bin as gain rises. Document the rig in the task.
-- [ ] **Closed-loop behaviour confirmed:** as driven gain rises the synthetic
+- [ ] **Closed-loop behavior confirmed:** as driven gain rises the synthetic
       peak grows; the app notches the correct band; after the notch the peak
       stops growing; the ramp resumes. Capture a run log.
 - [ ] **Safety asserted (the point of this task):**
@@ -41,7 +41,7 @@ See [`docs/ringout-design.md`](../docs/ringout-design.md) ("Safety").
 
 ## Notes
 
-- The emulator manages OSC state only (no audio), so the "feedback" is modelled
+- The emulator manages OSC state only (no audio), so the "feedback" is modeled
   in the test harness, not acoustically — sufficient to validate control logic,
   ramp, ceiling, abort, and profile handoff.
 - Real-PA acoustic validation is a separate, supervised field test — out of scope
diff --git a/tasks/T-027.md b/tasks/T-027.md
index 5082d4d..b98cee5 100644
--- a/tasks/T-027.md
+++ b/tasks/T-027.md
@@ -88,9 +88,9 @@ Host-testable: spot-check `RTA_BIN_FREQ[56] ≈ 1000`, monotonic increasing, and
   + `rta_bin_for_freq()`. `cli/test_rta_bins.c` — **10 tests green** (anchor,
   endpoints, monotonic, GEQ-anchors→ISO, inverse round-trip + clamps).
 - `cli/rta_probe.py` — live field probe (subscribe `/meters/4`, print peak bin
-  + modelled Hz).
+  + modeled Hz).
 - **Verified on a real XR18 (fw 1.20):** 1 kHz pistonphone on ch1 → **peak bin 56**
-  (−51 dBFS, neighbours 30+ dB down) = the table's anchor. ✓
+  (−51 dBFS, neighbors 30+ dB down) = the table's anchor. ✓
 - **`/meters/4` IS the RTA** — blob `[BE size][LE count=100][int16 LE ×100]`,
   matching `parse_meters4_blob`. The `/meters/5` question is moot.
 
diff --git a/tasks/T-029.md b/tasks/T-029.md
index b6929e4..2adf424 100644
--- a/tasks/T-029.md
+++ b/tasks/T-029.md
@@ -172,7 +172,7 @@ Confirm `make XAir_ToastSaver` still compiles clean.
         now inactive). Verifies the re-trigger path.
       - Mid-release: feed the spike back before `release_frames` → assert
         snap-back to `cut_val` (ramp_val == cut_val with release_hold == 0).
-      - Narrowness gate: spike with a broadband neighbour shape → no notch
+      - Narrowness gate: spike with a broadband neighbor shape → no notch
         (`confirm_hold` never increments).
       - `from_profile` band: mark band 5 as `from_profile`; feed its bin below
         threshold for `release_frames + 1` frames → assert no `TOAST_RELEASE_*`
diff --git a/tasks/T-033.md b/tasks/T-033.md
index 46056f0..5657165 100644
--- a/tasks/T-033.md
+++ b/tasks/T-033.md
@@ -54,7 +54,7 @@ therefore checks the FX rack slot, not a bus insert.
 --model {xr18,x32}   Mixer model (default: xr18)
 ```
 
-All existing behaviour is unchanged when `--model xr18` (or flag omitted).
+All existing behavior is unchanged when `--model xr18` (or flag omitted).
 
 ## Changes
 
@@ -65,7 +65,7 @@ All existing behaviour is unchanged when `--model xr18` (or flag omitted).
 | x32   | 10023 |
 
 ### RTA meter subscription (`osc_io.c`)
-Parameterise `osc_subscribe_meters4()` to accept the meter path:
+Parameterize `osc_subscribe_meters4()` to accept the meter path:
 
 ```c
 void osc_subscribe_meters(OscConn *c, const char *meter_path);
diff --git a/tasks/T-034.md b/tasks/T-034.md
index 7c95e86..33270fc 100644
--- a/tasks/T-034.md
+++ b/tasks/T-034.md
@@ -70,7 +70,7 @@ complexity 148 to ~30 and makes the reactive mode independently testable.
    The preflight `run_doctor` should stay inside `run_reactive` only — not run
    for `--ringout`.
 
-3. No behaviour changes. All flags, defaults, and flow remain identical. The
+3. No behavior changes. All flags, defaults, and flow remain identical. The
    only observable difference is that `osc_conn_close` for the ring-out path
    moves out of `run_ringout` and into `main()` (it is already there — check
    this, do not double-close).
diff --git a/tasks/T-035.md b/tasks/T-035.md
index f58845c..bb26cdc 100644
--- a/tasks/T-035.md
+++ b/tasks/T-035.md
@@ -95,7 +95,7 @@ All internal functions that currently take `AppState *s` become
 suffices, `(const AppConfig *cfg)`.
 
 `main()` declares `AppConfig cfg; AppRuntime rt; memset(&cfg, 0, ...);
-memset(&rt, 0, ...);` and initialises them separately.
+memset(&rt, 0, ...);` and initializes them separately.
 
 ## Mechanical substitution guide
 
@@ -135,17 +135,17 @@ Every static function that takes `AppState *s` needs its signature updated:
 - [ ] `make test-ringout-emu-x32` passes
 - [ ] `AppState` no longer exists in `XAir_ToastSaver.c`
 - [ ] `AppConfig` fields are never mutated after `main()` finishes parsing
-- [ ] No behaviour changes (no flag defaults changed, no logic changed)
+- [ ] No behavior changes (no flag defaults changed, no logic changed)
 
 ## Notes
 
 - `load_mic_cal` is called during arg parsing (from `getopt` case
-  `OPT_MIC_CAL`), before `AppRuntime` is fully initialised except for
+  `OPT_MIC_CAL`), before `AppRuntime` is fully initialized except for
   `mic_corr`, which is zeroed by `memset`. This is fine — just call
   `load_mic_cal(&rt, optarg)` during parsing.
 - `make_reactive_ctx` builds a `TuiReactiveCtx` from AppState fields — update
   it to read from both `cfg` and `rt`.
 - This is a large mechanical change. Go function-by-function in file order.
   Compile after each function update to catch typos early (`make XAir_ToastSaver`).
-- Do not change any logic, thresholds, flow, or external behaviour — only the
+- Do not change any logic, thresholds, flow, or external behavior — only the
   struct split and signature updates.
diff --git a/tasks/T-036.md b/tasks/T-036.md
index 14ec13d..dd16259 100644
--- a/tasks/T-036.md
+++ b/tasks/T-036.md
@@ -109,12 +109,12 @@ if (s->model->check_outputs) {
 - If T-035 has landed, the signatures use `(const AppConfig*, AppRuntime*, ...)`
   instead of `AppState*`. Update `check_outputs_fn` accordingly.
 - `xr18_check_outputs` should be defined immediately before `MODEL_XR18` in the
-  file so the function pointer is valid at static initialisation time — but note
-  that in C, static struct initialisers with function pointers are fine as long
+  file so the function pointer is valid at static initialization time — but note
+  that in C, static struct initializers with function pointers are fine as long
   as the function is declared before the struct literal. A forward declaration
   works too.
 - The X32 INFO message can stay inline in `run_doctor` (no vtable slot needed
   for the fallback message) — it is a general "routing is user responsibility"
-  note, not model-specific behaviour.
+  note, not model-specific behavior.
 - Do not change the FX slot check — it already works for both models and needs
   no per-model dispatch.
diff --git a/tasks/T-037.md b/tasks/T-037.md
index 425106f..899a025 100644
--- a/tasks/T-037.md
+++ b/tasks/T-037.md
@@ -103,7 +103,7 @@ XAir_ToastSaver: X32lib
 - [ ] All `g_log_scroll` mutations go through `log_scroll_down()` /
       `log_scroll_reset()` (search for bare `g_log_scroll` in .c — none should
       remain)
-- [ ] No behaviour change
+- [ ] No behavior change
 
 ## Notes
 
@@ -111,7 +111,7 @@ XAir_ToastSaver: X32lib
   `run_ringout` and `run_reactive`. The write path (`g_log_scroll++`) maps to
   `log_scroll_down()`. The reset (`g_log_scroll = 0`) can be handled in
   `log_push` internally (the current logic already does `if (g_log_scroll > 0)
-  g_log_scroll++` — keep this behaviour inside `log.c`).
+  g_log_scroll++` — keep this behavior inside `log.c`).
 - The accessors (`log_count`, `log_scroll`, `log_lines`) are intentionally
   simple getters — no locking needed (single-threaded).
 - Do not add a test binary for `log.c` in this task — the existing integration
diff --git a/tasks/T-038.md b/tasks/T-038.md
index c2c72e2..9abb305 100644
--- a/tasks/T-038.md
+++ b/tasks/T-038.md
@@ -124,7 +124,7 @@ Also rename the section comment from `/* ── /meters/4 blob handler */` to
 - [ ] `handle_calibration_frame` and `handle_detection_frame` are each ≤ 35
       lines
 - [ ] The dispatcher `handle_meters` is ≤ 8 lines
-- [ ] No behaviour change
+- [ ] No behavior change
 
 ## Notes