Add inline STT test panel and fix whisperx pyannote auth error message

- STT section now has a Quick test panel: select backend, hit mic button,
  see transcript. Records via MediaRecorder, posts to /api/transcribe-bytes.
- _transcribe_audio detects the whisperx-gpu 'NoneType/to' error (caused by
  pyannote/speaker-diarization-3.1 requiring a HuggingFace token) and
  replaces it with an actionable message explaining how to fix it.
- _to_wav_16k added for STT audio conversion (Whisper/wav2vec2 expect 16kHz).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

server.py

@@ -2104,6 +2104,12 @@ def _transcribe_audio(src: Path, settings: dict, backend: str = "configured") ->
             detail = resp.text[:300].strip()
         if not detail or detail.lower() in {"internal server error", "unknown error"}:
             detail = f"HTTP {resp.status_code} — backend may be misconfigured or missing CUDA support"
+        # Detect pyannote speaker-diarization gated-model auth error
+        if "'NoneType'" in detail and "'to'" in detail:
+            detail = ("Speaker diarization failed — pyannote/speaker-diarization-3.1 requires "
+                      "a HuggingFace token. Get one at hf.co/settings/tokens and accept the "
+                      "model license at hf.co/pyannote/speaker-diarization-3.1, then add the "
+                      "token to the whisperx-gpu container env as HF_TOKEN.")
         raise RuntimeError(f"STT ({stt_url}): {detail}")
     return _transcription_text_from_response(resp), backend
 

static/app.js

@@ -7455,6 +7455,88 @@ document.querySelectorAll('.dc-refresh-btn').forEach(b => b.addEventListener('cl
     });
   });
 
+  // ── Inline STT quick-test panel ─────────────────────────────────────────
+  (async function initSttTestPanel() {
+    const panel    = $('stt-test-panel');
+    const sel      = $('stt-test-backend');
+    const micBtn   = $('stt-test-mic-btn');
+    const statusEl = $('stt-test-status');
+    const resultEl = $('stt-test-result');
+    if (!panel || !micBtn) return;
+
+    // Populate backend dropdown
+    async function refreshSttTestBackends() {
+      try {
+        const d = await fetch('/api/stt-backends').then(r => r.json());
+        const prev = sel.value;
+        sel.innerHTML = (d.backends || []).map(b =>
+          `<option value="${escHtml(b.id)}"${!b.available ? ' disabled' : ''}>${b.available ? '✓' : '✗'} ${escHtml(b.label)}</option>`
+        ).join('');
+        if (prev && sel.querySelector(`option[value="${CSS.escape(prev)}"]`)) sel.value = prev;
+      } catch(_) {}
+    }
+    refreshSttTestBackends();
+
+    if (!navigator.mediaDevices?.getUserMedia) {
+      micBtn.disabled = true;
+      micBtn.title = 'Microphone unavailable (requires HTTPS or localhost)';
+      return;
+    }
+
+    let mediaRecorder = null;
+    let chunks = [];
+
+    micBtn.addEventListener('click', async () => {
+      if (mediaRecorder && mediaRecorder.state === 'recording') {
+        mediaRecorder.stop();
+        return;
+      }
+      if (micBtn.classList.contains('busy')) return;
+      chunks = [];
+      try {
+        const stream = await navigator.mediaDevices.getUserMedia({ audio: true });
+        mediaRecorder = new MediaRecorder(stream);
+        mediaRecorder.ondataavailable = e => { if (e.data.size) chunks.push(e.data); };
+        mediaRecorder.onstop = async () => {
+          stream.getTracks().forEach(t => t.stop());
+          micBtn.className = 'stt-test-mic busy';
+          micBtn.innerHTML = '<span class="mdi mdi-loading mdi-spin"></span>';
+          statusEl.textContent = 'Transcribing…';
+          resultEl.style.display = 'none';
+          const blob = new Blob(chunks, { type: 'audio/webm' });
+          const form = new FormData();
+          form.append('file', blob, 'audio.webm');
+          form.append('backend', sel.value || 'configured');
+          try {
+            const r = await fetch('/api/transcribe-bytes', { method: 'POST', body: form });
+            const d = await r.json();
+            if (d.text !== undefined) {
+              resultEl.textContent = d.text || '(no speech detected)';
+              statusEl.textContent = 'Done';
+            } else {
+              resultEl.textContent = '⚠ ' + (d.detail || JSON.stringify(d));
+              statusEl.textContent = 'Error';
+            }
+          } catch(e) {
+            resultEl.textContent = '⚠ ' + e.message;
+            statusEl.textContent = 'Error';
+          }
+          resultEl.style.display = '';
+          micBtn.className = 'stt-test-mic';
+          micBtn.innerHTML = '<span class="mdi mdi-microphone"></span>';
+        };
+        mediaRecorder.start();
+        micBtn.className = 'stt-test-mic recording';
+        micBtn.innerHTML = '<span class="mdi mdi-stop"></span>';
+        statusEl.textContent = 'Recording…';
+        resultEl.style.display = 'none';
+      } catch(e) {
+        statusEl.textContent = 'Mic error';
+        toast('Microphone error: ' + e.message, 'error');
+      }
+    });
+  })();
+
 })();
 
 // ── LLM snippet collapse ───────────────────────────────────────────────────

static/sections/s-llms.html

@@ -292,6 +292,19 @@ Check "Enable CORS" for browser access</pre>
 
         <div id="dc-grid-stt" class="dc-grid" style="margin-bottom:14px"></div>
 
+        <!-- Inline STT test panel -->
+        <div class="stt-test-panel" id="stt-test-panel">
+          <div class="stt-test-row">
+            <span class="stt-test-label">Quick test</span>
+            <select id="stt-test-backend" class="stt-test-select"></select>
+            <button id="stt-test-mic-btn" class="stt-test-mic" title="Hold to record, click again to stop">
+              <span class="mdi mdi-microphone"></span>
+            </button>
+            <span id="stt-test-status" class="stt-test-status">Ready</span>
+          </div>
+          <div id="stt-test-result" class="stt-test-result" style="display:none"></div>
+        </div>
+
         <h4 class="llm-local-cat">Other local STT</h4>
         <div class="llm-local-grid">
 

static/style.css

@@ -1706,6 +1706,73 @@ code { background: var(--panel); border-radius: 4px; padding: 1px 5px; font-fami
   grid-column: 1 / -1; padding: 18px 0; text-align: center;
   font-size: 13px; color: var(--subtext);
 }
+.stt-test-panel {
+  background: var(--surface2);
+  border: 1px solid var(--border);
+  border-radius: 10px;
+  padding: 10px 14px;
+  margin-bottom: 18px;
+  display: flex;
+  flex-direction: column;
+  gap: 8px;
+}
+.stt-test-row {
+  display: flex;
+  align-items: center;
+  gap: 10px;
+  flex-wrap: wrap;
+}
+.stt-test-label {
+  font-size: 12px;
+  font-weight: 600;
+  color: var(--subtext);
+  text-transform: uppercase;
+  letter-spacing: .04em;
+  white-space: nowrap;
+}
+.stt-test-select {
+  font-size: 13px;
+  padding: 4px 8px;
+  border-radius: 6px;
+  border: 1px solid var(--border);
+  background: var(--surface);
+  color: var(--text);
+  flex: 1;
+  min-width: 160px;
+  max-width: 320px;
+}
+.stt-test-mic {
+  width: 34px; height: 34px;
+  border-radius: 50%;
+  border: none;
+  background: var(--accent);
+  color: #fff;
+  font-size: 16px;
+  cursor: pointer;
+  display: flex; align-items: center; justify-content: center;
+  flex-shrink: 0;
+  transition: background .15s, transform .1s;
+}
+.stt-test-mic:hover { background: var(--accent-hover, #0060d0); }
+.stt-test-mic.recording { background: #e53e3e; animation: stt-pulse 1s infinite; }
+.stt-test-mic.busy { background: var(--subtext); cursor: not-allowed; }
+@keyframes stt-pulse { 0%,100% { transform: scale(1); } 50% { transform: scale(1.1); } }
+.stt-test-status {
+  font-size: 12px;
+  color: var(--subtext);
+  min-width: 60px;
+}
+.stt-test-result {
+  font-size: 13px;
+  color: var(--text);
+  background: var(--surface);
+  border: 1px solid var(--border);
+  border-radius: 6px;
+  padding: 8px 10px;
+  white-space: pre-wrap;
+  word-break: break-word;
+  min-height: 32px;
+}
 .dc-load-err { color: var(--red) !important; }
 .dc-card {
   background: var(--surface); border: 1px solid var(--border);