Add Pending_Query class and voice interaction improvements

- lib/pending-query.mjs: new state machine for query accumulation
  wake word, silence timer, send/cancel/pause/resume, instant dispatch,
  mode toggle (always listen / stop listening), mode query
- query-demo.mjs: refactored to use Pending_Query; wake word on by default
  with silence timer; chimes for dispatch/working/cancel/activate
- tts-server.mjs: track last_speak_at, expose /activity endpoint,
  chime playback via Python queue (soundfile + librosa), preload on startup
- chatterbox-server.py: chime and preload commands via stdin protocol
- lib/chatterbox-tts.mjs: play_chime and preload_chime methods
- test-chime.mjs: simple chime test script
- voices.yaml: configured ready/cancel/working/dispatch chimes
- CLEANUP-PLAN.md: updated with current state, command vocabulary, future plans

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

tts-server.mjs

@@ -20,7 +20,6 @@
 import * as http       from 'node:http'
 import * as fs         from 'node:fs'
 import * as path       from 'node:path'
-import { spawn }       from 'node:child_process'
 import yaml            from 'js-yaml'
 import { Chatterbox_Tts } from './lib/chatterbox-tts.mjs'
 
@@ -38,6 +37,7 @@ function reload_config() {
 
 let { voices, chimes } = reload_config()
 let current_voice = null  // name of active voice, or null
+let last_speak_at = 0
 
 // --- TTS setup ---
 const tts = new Chatterbox_Tts()
@@ -45,6 +45,16 @@ process.stderr.write('[tts-server] starting chatterbox...\n')
 await tts.init()
 process.stderr.write('[tts-server] chatterbox ready\n')
 
+// Preload all configured chimes so first play has no decode latency
+{
+	const { chimes: configured_chimes } = reload_config()
+	for (const [name, file] of Object.entries(configured_chimes)) {
+		tts.preload_chime(file).catch(err =>
+			process.stderr.write(`[tts-server] preload failed for chime '${name}': ${err.message}\n`)
+		)
+	}
+}
+
 // Serialize all speak requests through a promise chain
 let queue = Promise.resolve()
 
@@ -72,19 +82,16 @@ function send(res, status, body) {
 	res.end(payload)
 }
 
-function play_file(file_path) {
-	return new Promise((resolve, reject) => {
-		const player = spawn('pacat', ['--playback', file_path])
-		player.on('close', code => code === 0 ? resolve() : reject(new Error(`pacat exited ${code}`)))
-		player.on('error', reject)
-	})
-}
 
 const server = http.createServer(async (req, res) => {
 	if (req.method === 'GET' && req.url === '/health') {
 		return send(res, 200, { ok: true })
 	}
 
+	if (req.method === 'GET' && req.url === '/activity') {
+		return send(res, 200, { last_speak_at })
+	}
+
 	if (req.method === 'GET' && req.url === '/voices') {
 		({ voices, chimes } = reload_config())
 		const list = Object.entries(voices).map(([name, v]) => ({
@@ -120,7 +127,7 @@ const server = http.createServer(async (req, res) => {
 		const file = chimes[name] ?? null
 		if (!file) return send(res, 404, { error: `chime not found: ${name}` })
 		try {
-			await enqueue(() => play_file(file))
+			await enqueue(() => tts.play_chime(file))
 			return send(res, 200, { ok: true })
 		} catch (err) {
 			return send(res, 500, { error: err.message })
@@ -139,6 +146,7 @@ const server = http.createServer(async (req, res) => {
 		if (!text) {
 			return send(res, 400, { error: 'text required' })
 		}
+		last_speak_at = Date.now()
 
 		// Inject current voice as default audio_prompt if none provided
 		if (!opts.audio_prompt && current_voice && voices[current_voice]) {