Test 1: record 5s from USB mic with RMS meter (PA_ALSA_PLUGHW for resampling)

docs/superpowers/plans/2026-06-11-voice-assistant-prototypes.md

@@ -10,6 +10,8 @@
 
 **TDD note:** These are hardware-in-the-loop probes, not modules with unit tests. The "test" for each task is a manual run on the Pi against the spec's pass criteria. Every code-producing task ends with a deploy + run step that names the specific behaviour to look for.
 
+**Audio note:** The Pi's USB Speaker Phone only supports 48 kHz natively. We set `PA_ALSA_PLUGHW=1` at the top of every test's `main.py` (before importing sounddevice) so PortAudio opens via ALSA's `plughw:` device, which auto-resamples to the requested 16 kHz.
+
 **Spec:** `docs/superpowers/specs/2026-06-11-voice-assistant-prototypes-design.md`
 
 ---
@@ -230,6 +232,9 @@ Pass criteria for the record stage:
 - Level meter visibly responds to speech and to silence.
 - out.wav is written and non-empty.
 """
+import os
+os.environ.setdefault("PA_ALSA_PLUGHW", "1")
+
 import sys
 import time
 
@@ -422,6 +427,9 @@ This task validates that the model loads, the mic stream produces 1280-sample
 frames, and predict() returns a non-empty score dict. Detection + beep come
 in the next task.
 """
+import os
+os.environ.setdefault("PA_ALSA_PLUGHW", "1")
+
 import sys
 import time
 
@@ -534,6 +542,9 @@ Pass criteria (per spec):
 - During ~1 minute of unrelated speech, <= ~1 false positive.
 - Python process < 50% of one core (check with htop).
 """
+import os
+os.environ.setdefault("PA_ALSA_PLUGHW", "1")
+
 import sys
 import time