feat(voice): bracketed emotion tags, [잡음] for noise, spoken emotion, concise replies

- Brain persona now prefixes every reply with one bracketed emotion tag
  (e.g. [반가움], [궁금]) and keeps replies to one or two short sentences.
- Empty/unrecognised audio (silence/noise) is reported as reply "[잡음]" with
  no TTS playback instead of an empty reply.
- TTS speaks the bracketed emotion too: "[힘차게] 안녕!" is synthesised as
  "힘차게, 안녕!" via a leading-tag -> spoken-word transform.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

wsai/dashboard.py

@@ -18,6 +18,7 @@ from __future__ import annotations
 import json
 import logging
 import queue
+import re
 import threading
 from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
 
@@ -25,6 +26,21 @@ from .monitor import Monitor
 
 log = logging.getLogger("wsai.dashboard")
 
+_EMOTION_RE = re.compile(r"^\s*\[([^\]]+)\]\s*(.*)", re.S)
+
+
+def _speech_text(reply: str) -> str:
+    """Turn a reply like ``[힘차게] 안녕!`` into what the TTS should actually say.
+
+    The bracketed emotion is spoken too (per user request), so the leading tag
+    becomes a natural spoken word: ``[힘차게] 안녕!`` -> ``힘차게, 안녕!``. Replies
+    without a tag are spoken as-is."""
+    m = _EMOTION_RE.match(reply or "")
+    if not m:
+        return reply
+    emotion, rest = m.group(1).strip(), m.group(2).strip()
+    return f"{emotion}, {rest}" if rest else emotion
+
 
 def _make_handler(dash: "Dashboard"):
     monitor = dash.monitor
@@ -259,12 +275,14 @@ class Dashboard:
             heard = (self._submit(self.stt.transcribe(wav)) or "").strip()
             turn.heard(heard or "(빈 결과)")
             if not heard:
-                # Nothing recognised (silence/noise): skip the turn, tell the bot.
+                # Nothing recognised (silence/noise): mark it as [잡음] and skip
+                # the brain/TTS so the bot plays nothing back.
+                turn.replied("[잡음]")
                 turn.finish()
-                return {"heard": heard, "reply": "", "wav": b""}
+                return {"heard": heard, "reply": "[잡음]", "wav": b""}
             reply_text = self._think(heard)
             turn.replied(reply_text)
-            out_path = self._submit(self.tts.synth(reply_text))
+            out_path = self._submit(self.tts.synth(_speech_text(reply_text)))
             with open(out_path, "rb") as f:
                 reply_wav = f.read()
             ms = int((time.monotonic() - t0) * 1000)