@alexkroman1/aai-ui 5.9.0 → 5.10.0

package/dist/default-client/assets/{playback-processor-X5id-na1.js → playback-processor-GPTQTyza.js}

@@ -1,4 +1,4 @@
-import{i as e}from"./index-fLjOiK97.js";import{t}from"./_module-url-BX0RuRU2.js";var n=t(`
+import{i as e}from"./index-D_RNiTIh.js";import{t}from"./_module-url-BX0RuRU2.js";var n=t(`
 class PlaybackProcessor extends AudioWorkletProcessor {
   constructor(options) {
     super();
@@ -26,6 +26,11 @@ class PlaybackProcessor extends AudioWorkletProcessor {
     // playing instead of writing past the end and going silent.
     this.capacity = rate * 60;
     this.samples = new Float32Array(this.capacity);
+    // Cadence for the 'progress' report in process(), counted in samples so
+    // the hot path never reads a clock. Survives resetTurn(): the host clamps
+    // upward only, so a stale count costs at most one extra report.
+    this.reportIntervalSamples = Math.floor((rate * 500) / 1000);
+    this.sinceReportSamples = 0;
     // Platform endianness probe: the wire format is PCM16 little-endian, so
     // the Int16Array fast path in ingestBytes is only valid on LE hosts
     // (every shipping browser target; the DataView path is the fallback).
@@ -45,6 +50,10 @@ class PlaybackProcessor extends AudioWorkletProcessor {
         this.stopTurn('interrupt');
       } else if (d.event === 'done') {
         this.isDone = true;
+        // Echoed back on this turn's 'stop' so the host can tell WHICH turn
+        // drained: a stop already in flight when a barge-in lands would
+        // otherwise settle the next turn's done() the moment it arrives.
+        this.doneTurn = d.turn ?? null;
       }
     };
   }
@@ -53,6 +62,8 @@ class PlaybackProcessor extends AudioWorkletProcessor {
   // reallocating the sample buffer or re-instantiating the worklet.
   resetTurn() {
     this.isDone = false;
+    // Host turn id carried by this turn's 'done' (null until one arrives).
+    this.doneTurn = null;
     this.playing = false;
     // Whether any real audio has been rendered this turn. Separates a turn's
     // pre-roll (nothing to extrapolate from, and not a defect) from a
@@ -85,10 +96,12 @@ class PlaybackProcessor extends AudioWorkletProcessor {
   // Must NOT return false from process() — a processor that stops is dead
   // for good, forcing a new node (and buffer) per reply.
   // \`reason\` ('interrupt' | 'done') tells the host which turn boundary this
-  // stop belongs to: interrupt-stops are dropped host-side (flush() already
-  // settled that turn), so they can never resolve a later turn's done() early.
+  // stop belongs to — interrupt-stops are dropped host-side, flush() having
+  // already settled that turn — and \`turn\` names WHICH turn drained (the id
+  // this turn's 'done' carried), so a drain-stop still in flight when a
+  // barge-in lands cannot settle the next turn's done() early.
   stopTurn(reason) {
-    this.port.postMessage({ event: 'stop', reason, stats: this.stats });
+    this.port.postMessage({ event: 'stop', reason, turn: this.doneTurn, stats: this.stats });
     this.resetTurn();
   }
 
@@ -225,6 +238,26 @@ class PlaybackProcessor extends AudioWorkletProcessor {
     const out = outputs[0][0];
     const avail = this.writePos - this.readPos;
 
+    // Report the unplayed backlog to the host (\`playback_progress\`). This IS
+    // the closed-loop signal: the host otherwise models playback open-loop —
+    // every forwarded chunk assumed to start playing on arrival at exactly
+    // 1.0x — and cannot see a buffer that has run ahead of the wall clock.
+    // Measured against a client draining at 0.6x, the host declared the line
+    // silent while it still held seconds of the reply, and then opened the
+    // speaking-edge gate over speech the caller had not heard.
+    //
+    // Sent from process() rather than a timer because this is the only place
+    // that sees the buffer on the audio thread; the counter is in QUANTA, so
+    // it costs an integer compare per render and no clock read. Silence is not
+    // reported: an empty buffer is what the host already assumes.
+    this.sinceReportSamples += out.length;
+    if (this.sinceReportSamples >= this.reportIntervalSamples) {
+      this.sinceReportSamples = 0;
+      if (avail > 0) {
+        this.port.postMessage({ event: 'progress', bufferedMs: (avail / sampleRate) * 1000 });
+      }
+    }
+
     // Filling: wait for the target. 'done' short-circuits it — what is
     // buffered is all there will be, so there is nothing left to wait for.
     if (!this.playing) {

package/dist/default-client/index.html

@@ -6,8 +6,8 @@
     <title>aai</title>
     <link rel="icon" href="./favicon.ico" />
     <style>html, body { background: #FBF8F2; margin: 0; }</style>
-    <script type="module" crossorigin src="./assets/index-fLjOiK97.js"></script>
-    <link rel="stylesheet" crossorigin href="./assets/index-CuPmIAbL.css">
+    <script type="module" crossorigin src="./assets/index-D_RNiTIh.js"></script>
+    <link rel="stylesheet" crossorigin href="./assets/index-1JIz81CD.css">
   </head>
   <body>
     <main id="app"></main>

package/dist/define-client.js

@@ -1,4 +1,4 @@
-import { r as fetchClientConfig, t as createSessionCore } from "./session-core-DqibBzJz.js";
+import { r as fetchClientConfig, t as createSessionCore } from "./session-core-CS37WE1G.js";
 import { SessionProvider, ThemeProvider } from "./context.js";
 import { t as ChatView } from "./chat-view-DFBmX5eX.js";
 import { t as pageBaseUrl } from "./_utils-CsqbIVGI.js";

package/dist/index.js

@@ -1,4 +1,4 @@
-import { i as loadClientConfig, n as buildAgentUrl, r as fetchClientConfig, t as createSessionCore } from "./session-core-DqibBzJz.js";
+import { i as loadClientConfig, n as buildAgentUrl, r as fetchClientConfig, t as createSessionCore } from "./session-core-CS37WE1G.js";
 import { SessionProvider, ThemeProvider, useSession, useSessionSelector, useTheme } from "./context.js";
 import { Button } from "./components/button.js";
 import { t as ChatView } from "./chat-view-DFBmX5eX.js";

package/dist/{session-core-DqibBzJz.js → session-core-CS37WE1G.js}

@@ -136,6 +136,15 @@ async function initAudioCapture(conn, msg, deps) {
 					console.debug("[aai-ui] sendAudio dropped: connection closed");
 				}
 			},
+			onPlaybackProgress: (bufferedMs) => {
+				if (!conn.generation.isCurrent(gen)) return;
+				try {
+					deps.sendJson({
+						type: "playback_progress",
+						bufferedMs
+					});
+				} catch {}
+			},
 			onError: (err) => {
 				if (!conn.generation.isCurrent(gen)) return;
 				reportAudioFailure(err.message);
@@ -207,14 +216,14 @@ function appendCapped(list, item, cap) {
 /**
 * Create the server→client message handlers for one session core.
 *
-* Encapsulates the two turn-boundary counters (`turnEpoch` for
-* discarding stale async audio completions, `customEventSeq` for event
-* dedup) that previously lived as closure locals in `createSessionCore`.
+* Encapsulates the per-session dedup counters (`customEventSeq`,
+* `messageSeq`, `toolCallSeq`) that previously lived as closure locals in
+* `createSessionCore`. The turn-boundary epoch is NOT one of them: it lives
+* on `conn` because the session core bumps it on teardown too (see
+* `ConnState.turn`).
 */
 function createMessageHandlers(deps) {
 	const { getSnapshot, updateState, conn, cleanupAudio } = deps;
-	/** Bumped on each turn boundary -- stale async callbacks check against this. */
-	const turnEpoch = createEpoch();
 	/** Monotonically increasing counter for custom events -- used by useEvent to deduplicate. */
 	let customEventSeq = 0;
 	/** Monotonically increasing counter for chat messages -- stable render keys
@@ -231,7 +240,7 @@ function createMessageHandlers(deps) {
 		}, MAX_CUSTOM_EVENTS) });
 	}
 	function handleUserTranscriptEvent(text) {
-		turnEpoch.bump();
+		conn.turn.bump();
 		updateState({
 			userTranscript: null,
 			messages: appendCapped(getSnapshot().messages, {
@@ -243,11 +252,15 @@ function createMessageHandlers(deps) {
 		});
 	}
 	/**
-	* `agent_transcript` carries the reply's text *so far* and is cumulative
-	* within a reply (see the protocol schema), so it renders as the live
-	* assistant bubble and only becomes a message when the reply closes. Pipeline
-	* mode sends one per piece of speech, so appending each would break a single
-	* reply into a message per sentence.
+	* `agent_transcript` carries the reply's text so far as a full-replacement
+	* snapshot (see the protocol schema), so it renders as the live assistant
+	* bubble and only becomes a message when the reply closes. Pipeline mode sends
+	* one per piece of speech, so appending each would break a single reply into a
+	* message per sentence.
+	*
+	* Replace rather than merge: a snapshot is not append-only. A pipeline reply's
+	* closing snapshot drops the dead-air filler the interim ones carried, so it
+	* can be shorter than its predecessor and differ mid-string.
 	*/
 	function handleAgentTranscriptEvent(text) {
 		updateState({ agentTranscript: text });
@@ -350,7 +363,7 @@ function createMessageHandlers(deps) {
 				updateState({ state: "listening" });
 				break;
 			case "cancelled":
-				turnEpoch.bump();
+				conn.turn.bump();
 				conn.voiceIO?.flush();
 				commitAgentTranscript();
 				updateState({
@@ -359,7 +372,7 @@ function createMessageHandlers(deps) {
 				});
 				break;
 			case "reset":
-				turnEpoch.bump();
+				conn.turn.bump();
 				conn.voiceIO?.flush();
 				updateState({
 					...CLEARED_SESSION_STATE,
@@ -390,12 +403,13 @@ function createMessageHandlers(deps) {
 		else if (conn.preInitAudio.length < MAX_PREINIT_AUDIO_CHUNKS) conn.preInitAudio.push(chunk);
 	}
 	/** See {@link MessageHandlers.settleWhenAudioDrained}. Captures
-	*  `turnEpoch` so a completion (or failure) that lands after a turn
-	*  boundary is discarded instead of overwriting the newer turn's state. */
+	*  `conn.turn` so a completion (or failure) that lands after a turn
+	*  boundary — including an audio-path teardown — is discarded instead of
+	*  overwriting the newer turn's (or the dead session's) state. */
 	function settleWhenAudioDrained(io) {
-		const gen = turnEpoch.current();
+		const gen = conn.turn.current();
 		io.done().then(() => {
-			if (!turnEpoch.isCurrent(gen)) return;
+			if (!conn.turn.isCurrent(gen)) return;
 			updateState({ state: "listening" });
 		}).catch((err) => {
 			console.warn("Audio playback done failed:", err);
@@ -404,7 +418,7 @@ function createMessageHandlers(deps) {
 	/**
 	* Signal that the server has finished sending audio for this turn.
 	* Waits for the audio queue to drain, then transitions state to `"listening"`.
-	* Uses the `turnEpoch` epoch to discard stale completions from interrupted turns.
+	* Uses `conn.turn` to discard stale completions from interrupted turns.
 	*/
 	function playAudioDone() {
 		const io = conn.voiceIO;
@@ -598,6 +612,7 @@ function createSessionCore(options) {
 		voiceIO: null,
 		audioSetupInFlight: false,
 		generation: createEpoch(),
+		turn: createEpoch(),
 		preInitAudio: [],
 		preInitDone: false
 	};
@@ -623,6 +638,7 @@ function createSessionCore(options) {
 	let serverIsBroker;
 	function cleanupAudio() {
 		conn.audioSetupInFlight = false;
+		conn.turn.bump();
 		conn.voiceIO?.close().catch(() => {});
 		conn.voiceIO = null;
 		conn.preInitAudio = [];
@@ -775,11 +791,13 @@ function createSessionCore(options) {
 	}
 	function cancel() {
 		if (!conn.ws || conn.ws.readyState !== WS_OPEN) return;
+		conn.turn.bump();
 		conn.voiceIO?.flush();
 		updateState({ state: "listening" });
 		sendJson({ type: "cancel" });
 	}
 	function reset() {
+		conn.turn.bump();
 		conn.voiceIO?.flush();
 		if (conn.ws && conn.ws.readyState === WS_OPEN) {
 			sendJson({ type: "reset" });

package/dist/session-core-messages.d.ts

@@ -51,9 +51,11 @@ type MessageHandlers = {
 /**
  * Create the server→client message handlers for one session core.
  *
- * Encapsulates the two turn-boundary counters (`turnEpoch` for
- * discarding stale async audio completions, `customEventSeq` for event
- * dedup) that previously lived as closure locals in `createSessionCore`.
+ * Encapsulates the per-session dedup counters (`customEventSeq`,
+ * `messageSeq`, `toolCallSeq`) that previously lived as closure locals in
+ * `createSessionCore`. The turn-boundary epoch is NOT one of them: it lives
+ * on `conn` because the session core bumps it on teardown too (see
+ * `ConnState.turn`).
  */
 export declare function createMessageHandlers(deps: MessageHandlerDeps): MessageHandlers;
 export {};

package/dist/session-core-types.d.ts

@@ -157,6 +157,19 @@ export type ConnState = {
      *  Prevents a stale initAudioCapture from assigning its voiceIO to a newer
      *  connection. */
     generation: Epoch;
+    /**
+     * Turn epoch, bumped at every turn boundary — a committed user turn, a
+     * barge-in, a reset, AND every audio-path teardown (`cleanupAudio`).
+     *
+     * Async playback completions (`settleWhenAudioDrained`) capture it and
+     * discard themselves when it has moved. The teardown bumps are why it lives
+     * here rather than as a closure local in the message handlers: a reply's
+     * drain resolves whenever the AudioContext closes, so a session torn down
+     * mid-speech (hang up, fatal error, reconnect) would otherwise have
+     * `state: "listening"` written over its "disconnected"/"error" a moment
+     * later — a dead session reporting a live mic.
+     */
+    turn: Epoch;
     /** Audio chunks that arrived before `voiceIO` was initialized — drained into
      *  the playback worklet once init completes. Closes the race between the
      *  server starting greeting audio (immediately on S2S connect) and the

package/dist/session-core.js

@@ -1,2 +1,2 @@
-import { t as createSessionCore } from "./session-core-DqibBzJz.js";
+import { t as createSessionCore } from "./session-core-CS37WE1G.js";
 export { createSessionCore };

package/dist/types.d.ts

@@ -1,6 +1,6 @@
 import type { DefaultToolResult } from "@alexkroman1/aai";
 import type { SessionErrorCode } from "@alexkroman1/aai/protocol";
-export { CAPTURE_STOP_ACK_TIMEOUT_MS, MIC_BUFFER_SECONDS, MIC_SEND_MAX_BUFFERED_BYTES, MIC_SILENCE_PROBE_MS, PLAYBACK_BUFFER_SECONDS, PLAYBACK_CONCEAL_FADE_MS, PLAYBACK_CONCEAL_FLOOR, PLAYBACK_DONE_MAX_WAIT_MS, PLAYBACK_DONE_POLL_MS, PLAYBACK_JITTER_MS, PLAYBACK_REFILL_MS, } from "@alexkroman1/aai";
+export { CAPTURE_STOP_ACK_TIMEOUT_MS, MIC_BUFFER_SECONDS, MIC_SEND_MAX_BUFFERED_BYTES, MIC_SILENCE_PROBE_MS, PLAYBACK_BUFFER_SECONDS, PLAYBACK_CONCEAL_FADE_MS, PLAYBACK_CONCEAL_FLOOR, PLAYBACK_DONE_MAX_WAIT_MS, PLAYBACK_DONE_POLL_MS, PLAYBACK_JITTER_MS, PLAYBACK_PROGRESS_INTERVAL_MS, PLAYBACK_REFILL_MS, } from "@alexkroman1/aai";
 /**
  * `getUserMedia` audio constraints for every capture path in this package.
  *

package/dist/types.js

@@ -1,4 +1,4 @@
-import { CAPTURE_STOP_ACK_TIMEOUT_MS, MIC_BUFFER_SECONDS, MIC_SEND_MAX_BUFFERED_BYTES, MIC_SILENCE_PROBE_MS, PLAYBACK_BUFFER_SECONDS, PLAYBACK_CONCEAL_FADE_MS, PLAYBACK_CONCEAL_FLOOR, PLAYBACK_DONE_MAX_WAIT_MS, PLAYBACK_DONE_POLL_MS, PLAYBACK_JITTER_MS, PLAYBACK_REFILL_MS } from "@alexkroman1/aai";
+import { CAPTURE_STOP_ACK_TIMEOUT_MS, MIC_BUFFER_SECONDS, MIC_SEND_MAX_BUFFERED_BYTES, MIC_SILENCE_PROBE_MS, PLAYBACK_BUFFER_SECONDS, PLAYBACK_CONCEAL_FADE_MS, PLAYBACK_CONCEAL_FLOOR, PLAYBACK_DONE_MAX_WAIT_MS, PLAYBACK_DONE_POLL_MS, PLAYBACK_JITTER_MS, PLAYBACK_PROGRESS_INTERVAL_MS, PLAYBACK_REFILL_MS } from "@alexkroman1/aai";
 //#region types.ts
 /**
 * `getUserMedia` audio constraints for every capture path in this package.
@@ -32,4 +32,4 @@ const VOICE_CAPTURE_CONSTRAINTS = {
 	voiceIsolation: false
 };
 //#endregion
-export { CAPTURE_STOP_ACK_TIMEOUT_MS, MIC_BUFFER_SECONDS, MIC_SEND_MAX_BUFFERED_BYTES, MIC_SILENCE_PROBE_MS, PLAYBACK_BUFFER_SECONDS, PLAYBACK_CONCEAL_FADE_MS, PLAYBACK_CONCEAL_FLOOR, PLAYBACK_DONE_MAX_WAIT_MS, PLAYBACK_DONE_POLL_MS, PLAYBACK_JITTER_MS, PLAYBACK_REFILL_MS, VOICE_CAPTURE_CONSTRAINTS };
+export { CAPTURE_STOP_ACK_TIMEOUT_MS, MIC_BUFFER_SECONDS, MIC_SEND_MAX_BUFFERED_BYTES, MIC_SILENCE_PROBE_MS, PLAYBACK_BUFFER_SECONDS, PLAYBACK_CONCEAL_FADE_MS, PLAYBACK_CONCEAL_FLOOR, PLAYBACK_DONE_MAX_WAIT_MS, PLAYBACK_DONE_POLL_MS, PLAYBACK_JITTER_MS, PLAYBACK_PROGRESS_INTERVAL_MS, PLAYBACK_REFILL_MS, VOICE_CAPTURE_CONSTRAINTS };

package/dist/worklets/playback-processor.d.ts

@@ -1,4 +1,4 @@
 /** Raw worklet source — exported so tests can evaluate the processor directly. */
-export declare const playbackProcessorSource = "\nclass PlaybackProcessor extends AudioWorkletProcessor {\n  constructor(options) {\n    super();\n    const opts = options.processorOptions || {};\n    // The node's context IS the playback context (audio.ts asserts its rate),\n    // so the worklet-global sampleRate is authoritative; the option exists\n    // for the node-less test harness.\n    const rate = opts.sampleRate ?? sampleRate;\n    // Fill target for the start of a turn. If 'done' arrives first (short\n    // utterance), start immediately instead of waiting for audio that is\n    // never coming.\n    this.jitterSamples = Math.floor((rate * (opts.jitterMs ?? 400)) / 1000);\n    // Fill target after an underrun \u2014 see PLAYBACK_REFILL_MS.\n    this.refillSamples = Math.floor((rate * (opts.refillMs ?? 200)) / 1000);\n    // Concealment source: a ring of the most recently played samples, looped\n    // under a decaying gain to cover a gap. Sized to the fade window, with a\n    // per-sample decay that reaches the floor exactly at its end.\n    this.concealCapacity = Math.max(1, Math.floor((rate * 40) / 1000));\n    this.concealBuf = new Float32Array(this.concealCapacity);\n    this.concealDecay = Math.exp(Math.log(0.001) / this.concealCapacity);\n    // Float32 ring buffer \u2014 PLAYBACK_BUFFER_SECONDS at the context sample\n    // rate. Allocated once for the node's lifetime; per-turn state resets via\n    // resetTurn(). writePos and readPos are absolute (monotonic) sample\n    // counts; the buffer is indexed modulo capacity so a longer reply keeps\n    // playing instead of writing past the end and going silent.\n    this.capacity = rate * 60;\n    this.samples = new Float32Array(this.capacity);\n    // Platform endianness probe: the wire format is PCM16 little-endian, so\n    // the Int16Array fast path in ingestBytes is only valid on LE hosts\n    // (every shipping browser target; the DataView path is the fallback).\n    this.littleEndian = new Uint8Array(new Uint16Array([1]).buffer)[0] === 1;\n    this.resetTurn();\n\n    this.port.onmessage = (e) => {\n      const d = e.data;\n      if (d.event === 'write') {\n        this.ingestBytes(d.buffer);\n      } else if (d.event === 'interrupt') {\n        // Applied eagerly, not deferred to the next process(): onmessage and\n        // process() never interleave (one audio thread), and a deferred\n        // interrupt would let 'write'/'done' frames for the NEXT turn\n        // coalesce in ahead of it and be wiped by the reset along with the\n        // cancelled turn's audio.\n        this.stopTurn('interrupt');\n      } else if (d.event === 'done') {\n        this.isDone = true;\n      }\n    };\n  }\n\n  // Reset per-turn state so the node is reusable across replies without\n  // reallocating the sample buffer or re-instantiating the worklet.\n  resetTurn() {\n    this.isDone = false;\n    this.playing = false;\n    // Whether any real audio has been rendered this turn. Separates a turn's\n    // pre-roll (nothing to extrapolate from, and not a defect) from a\n    // mid-turn underrun.\n    this.hasPlayed = false;\n    this.fillTarget = this.jitterSamples;\n    // Carry-over byte for split samples across chunks\n    this.carry = null;\n    this.writePos = 0;\n    this.readPos = 0;\n    // Concealment ring state and the current fade position.\n    this.concealLen = 0;\n    this.concealWrite = 0;\n    this.concealPos = 0;\n    this.concealGain = 1;\n    // Episode flags, so a multi-quantum gap counts as one event.\n    this.concealing = false;\n    this.concealedSilence = false;\n    // Reported to the host on 'stop'. A fresh object per turn: the one just\n    // posted must not be mutated by the next turn.\n    this.stats = {\n      concealedSamples: 0,\n      silentConcealedSamples: 0,\n      concealmentEvents: 0,\n      silentConcealmentEvents: 0,\n    };\n  }\n\n  // End the current turn: notify the host and rearm for the next reply.\n  // Must NOT return false from process() \u2014 a processor that stops is dead\n  // for good, forcing a new node (and buffer) per reply.\n  // `reason` ('interrupt' | 'done') tells the host which turn boundary this\n  // stop belongs to: interrupt-stops are dropped host-side (flush() already\n  // settled that turn), so they can never resolve a later turn's done() early.\n  stopTurn(reason) {\n    this.port.postMessage({ event: 'stop', reason, stats: this.stats });\n    this.resetTurn();\n  }\n\n  // Cover a quantum (from `start`) where real audio should have been.\n  //\n  // Before the turn's first samples there is nothing to extrapolate from, so\n  // the gap is plain silence and counted as nothing \u2014 WebRTC likewise only\n  // counts concealment once playout has begun. After that, loop the retained\n  // tail under a decaying gain: a hard zero-fill is a discontinuity mid-word,\n  // which is the click that makes a brief stall sound like breakage.\n  coverGap(out, start) {\n    if (!this.hasPlayed) {\n      out.fill(0, start);\n      return;\n    }\n    if (!this.concealing) {\n      this.concealing = true;\n      this.concealedSilence = false;\n      this.stats.concealmentEvents++;\n    }\n    const total = out.length - start;\n    const len = this.concealLen;\n    let silent = 0;\n    // Second condition: once the fade has decayed to the floor, every sample\n    // the loop would emit is 0 anyway \u2014 bulk-fill instead of running 128\n    // branchy iterations per quantum for the whole tail of a long stall.\n    if (len === 0 || this.concealGain < 0.001) {\n      out.fill(0, start);\n      silent = total;\n    } else {\n      let g = this.concealGain;\n      for (let i = start; i < out.length; i++) {\n        if (g < 0.001) {\n          // The fade has run out: keep counting the gap, but stop looping a\n          // fragment that is now inaudible anyway.\n          out[i] = 0;\n          silent++;\n          continue;\n        }\n        out[i] = this.concealBuf[this.concealPos] * g;\n        this.concealPos = this.concealPos + 1 === len ? 0 : this.concealPos + 1;\n        g *= this.concealDecay;\n      }\n      this.concealGain = g;\n    }\n    this.stats.concealedSamples += total;\n    if (silent > 0) {\n      this.stats.silentConcealedSamples += silent;\n      if (!this.concealedSilence) {\n        this.concealedSilence = true;\n        this.stats.silentConcealmentEvents++;\n      }\n    }\n  }\n\n  // Retain the tail of a rendered quantum as the next gap's concealment\n  // source, and close any episode the real audio just ended.\n  rememberTail(out, n) {\n    const cap = this.concealCapacity;\n    const take = Math.min(n, cap);\n    // Bulk copies (this runs on every cleanly rendered quantum): the tail is\n    // one contiguous source run, landing in at most two ring runs.\n    const tail = out.subarray(n - take, n);\n    const first = Math.min(take, cap - this.concealWrite);\n    this.concealBuf.set(tail.subarray(0, first), this.concealWrite);\n    if (take > first) this.concealBuf.set(tail.subarray(first), 0);\n    this.concealWrite = (this.concealWrite + take) % cap;\n    this.concealLen = Math.min(cap, this.concealLen + take);\n    // Read the loop oldest-first; once the ring is full the write cursor is\n    // the oldest retained sample.\n    this.concealPos = this.concealLen === cap ? this.concealWrite : 0;\n    this.concealing = false;\n    this.concealGain = 1;\n  }\n\n  ingestBytes(uint8) {\n    let bytes = uint8;\n\n    if (this.carry !== null) {\n      const merged = new Uint8Array(1 + bytes.length);\n      merged[0] = this.carry;\n      merged.set(bytes, 1);\n      bytes = merged;\n      this.carry = null;\n    }\n\n    if (bytes.length % 2 !== 0) {\n      this.carry = bytes[bytes.length - 1];\n      bytes = bytes.subarray(0, bytes.length - 1);\n    }\n\n    if (bytes.length === 0) return;\n    const numSamples = bytes.length / 2;\n    const cap = this.capacity;\n    const samples = this.samples;\n    if (this.littleEndian && (bytes.byteOffset & 1) === 0) {\n      // Fast path: 2-byte-aligned LE bytes wrap directly as an Int16Array;\n      // copy wrap-aware in at most two runs with no per-sample DataView call\n      // or modulo. This runs on the realtime audio thread.\n      const int16 = new Int16Array(bytes.buffer, bytes.byteOffset, numSamples);\n      let src = 0;\n      let dst = this.writePos % cap;\n      while (src < numSamples) {\n        const run = Math.min(numSamples - src, cap - dst);\n        for (let j = 0; j < run; j++) {\n          samples[dst + j] = int16[src + j] / 0x8000;\n        }\n        src += run;\n        dst = 0;\n      }\n    } else {\n      // Odd byte offset (or big-endian host): fall back to per-sample reads.\n      const view = new DataView(bytes.buffer, bytes.byteOffset, bytes.length);\n      let dst = this.writePos % cap;\n      for (let i = 0; i < numSamples; i++) {\n        samples[dst] = view.getInt16(i * 2, true) / 0x8000;\n        dst++;\n        if (dst === cap) dst = 0;\n      }\n    }\n    this.writePos += numSamples;\n    // If the producer outran the consumer by more than the buffer holds, drop\n    // the oldest unplayed audio rather than reading samples we've overwritten.\n    if (this.writePos - this.readPos > this.capacity) {\n      this.readPos = this.writePos - this.capacity;\n    }\n  }\n\n  process(inputs, outputs) {\n    // No output wired up yet \u2014 nothing to render this quantum. Throwing here\n    // would permanently kill the processor (the node is persistent per\n    // session), so guard like the capture processor does.\n    if (!outputs[0] || !outputs[0][0]) return true;\n    const out = outputs[0][0];\n    const avail = this.writePos - this.readPos;\n\n    // Filling: wait for the target. 'done' short-circuits it \u2014 what is\n    // buffered is all there will be, so there is nothing left to wait for.\n    if (!this.playing) {\n      if (avail >= this.fillTarget || this.isDone) {\n        this.playing = true;\n      } else {\n        this.coverGap(out, 0);\n        return true;\n      }\n    }\n\n    // Underrun: this quantum cannot be filled and more audio is still coming.\n    // Go back to filling (at the refill target) and cover the gap, leaving\n    // readPos untouched \u2014 the fragment stays buffered and plays intact once\n    // the buffer recovers, instead of being dribbled out a few samples at a\n    // time for the rest of the turn.\n    if (avail < out.length && !this.isDone) {\n      this.playing = false;\n      this.fillTarget = this.refillSamples;\n      this.coverGap(out, 0);\n      return true;\n    }\n\n    if (avail > 0) {\n      const n = Math.min(avail, out.length);\n      // Copy from the ring buffer, splitting across the wrap boundary.\n      const start = this.readPos % this.capacity;\n      const first = Math.min(n, this.capacity - start);\n      out.set(this.samples.subarray(start, start + first), 0);\n      if (n > first) out.set(this.samples.subarray(0, n - first), first);\n      this.readPos += n;\n      // Only reachable with n < out.length on the turn's final partial\n      // quantum (the underrun branch above catches every other case).\n      out.fill(0, n);\n      this.hasPlayed = true;\n      this.rememberTail(out, n);\n      return true;\n    }\n\n    // Drained and done: end the turn. Not reachable mid-turn \u2014 an empty\n    // buffer with audio still coming is the underrun branch above.\n    out.fill(0);\n    if (this.isDone) {\n      this.stopTurn('done');\n    }\n    return true;\n  }\n}\n\nregisterProcessor('playback-processor', PlaybackProcessor);\n";
+export declare const playbackProcessorSource = "\nclass PlaybackProcessor extends AudioWorkletProcessor {\n  constructor(options) {\n    super();\n    const opts = options.processorOptions || {};\n    // The node's context IS the playback context (audio.ts asserts its rate),\n    // so the worklet-global sampleRate is authoritative; the option exists\n    // for the node-less test harness.\n    const rate = opts.sampleRate ?? sampleRate;\n    // Fill target for the start of a turn. If 'done' arrives first (short\n    // utterance), start immediately instead of waiting for audio that is\n    // never coming.\n    this.jitterSamples = Math.floor((rate * (opts.jitterMs ?? 400)) / 1000);\n    // Fill target after an underrun \u2014 see PLAYBACK_REFILL_MS.\n    this.refillSamples = Math.floor((rate * (opts.refillMs ?? 200)) / 1000);\n    // Concealment source: a ring of the most recently played samples, looped\n    // under a decaying gain to cover a gap. Sized to the fade window, with a\n    // per-sample decay that reaches the floor exactly at its end.\n    this.concealCapacity = Math.max(1, Math.floor((rate * 40) / 1000));\n    this.concealBuf = new Float32Array(this.concealCapacity);\n    this.concealDecay = Math.exp(Math.log(0.001) / this.concealCapacity);\n    // Float32 ring buffer \u2014 PLAYBACK_BUFFER_SECONDS at the context sample\n    // rate. Allocated once for the node's lifetime; per-turn state resets via\n    // resetTurn(). writePos and readPos are absolute (monotonic) sample\n    // counts; the buffer is indexed modulo capacity so a longer reply keeps\n    // playing instead of writing past the end and going silent.\n    this.capacity = rate * 60;\n    this.samples = new Float32Array(this.capacity);\n    // Cadence for the 'progress' report in process(), counted in samples so\n    // the hot path never reads a clock. Survives resetTurn(): the host clamps\n    // upward only, so a stale count costs at most one extra report.\n    this.reportIntervalSamples = Math.floor((rate * 500) / 1000);\n    this.sinceReportSamples = 0;\n    // Platform endianness probe: the wire format is PCM16 little-endian, so\n    // the Int16Array fast path in ingestBytes is only valid on LE hosts\n    // (every shipping browser target; the DataView path is the fallback).\n    this.littleEndian = new Uint8Array(new Uint16Array([1]).buffer)[0] === 1;\n    this.resetTurn();\n\n    this.port.onmessage = (e) => {\n      const d = e.data;\n      if (d.event === 'write') {\n        this.ingestBytes(d.buffer);\n      } else if (d.event === 'interrupt') {\n        // Applied eagerly, not deferred to the next process(): onmessage and\n        // process() never interleave (one audio thread), and a deferred\n        // interrupt would let 'write'/'done' frames for the NEXT turn\n        // coalesce in ahead of it and be wiped by the reset along with the\n        // cancelled turn's audio.\n        this.stopTurn('interrupt');\n      } else if (d.event === 'done') {\n        this.isDone = true;\n        // Echoed back on this turn's 'stop' so the host can tell WHICH turn\n        // drained: a stop already in flight when a barge-in lands would\n        // otherwise settle the next turn's done() the moment it arrives.\n        this.doneTurn = d.turn ?? null;\n      }\n    };\n  }\n\n  // Reset per-turn state so the node is reusable across replies without\n  // reallocating the sample buffer or re-instantiating the worklet.\n  resetTurn() {\n    this.isDone = false;\n    // Host turn id carried by this turn's 'done' (null until one arrives).\n    this.doneTurn = null;\n    this.playing = false;\n    // Whether any real audio has been rendered this turn. Separates a turn's\n    // pre-roll (nothing to extrapolate from, and not a defect) from a\n    // mid-turn underrun.\n    this.hasPlayed = false;\n    this.fillTarget = this.jitterSamples;\n    // Carry-over byte for split samples across chunks\n    this.carry = null;\n    this.writePos = 0;\n    this.readPos = 0;\n    // Concealment ring state and the current fade position.\n    this.concealLen = 0;\n    this.concealWrite = 0;\n    this.concealPos = 0;\n    this.concealGain = 1;\n    // Episode flags, so a multi-quantum gap counts as one event.\n    this.concealing = false;\n    this.concealedSilence = false;\n    // Reported to the host on 'stop'. A fresh object per turn: the one just\n    // posted must not be mutated by the next turn.\n    this.stats = {\n      concealedSamples: 0,\n      silentConcealedSamples: 0,\n      concealmentEvents: 0,\n      silentConcealmentEvents: 0,\n    };\n  }\n\n  // End the current turn: notify the host and rearm for the next reply.\n  // Must NOT return false from process() \u2014 a processor that stops is dead\n  // for good, forcing a new node (and buffer) per reply.\n  // `reason` ('interrupt' | 'done') tells the host which turn boundary this\n  // stop belongs to \u2014 interrupt-stops are dropped host-side, flush() having\n  // already settled that turn \u2014 and `turn` names WHICH turn drained (the id\n  // this turn's 'done' carried), so a drain-stop still in flight when a\n  // barge-in lands cannot settle the next turn's done() early.\n  stopTurn(reason) {\n    this.port.postMessage({ event: 'stop', reason, turn: this.doneTurn, stats: this.stats });\n    this.resetTurn();\n  }\n\n  // Cover a quantum (from `start`) where real audio should have been.\n  //\n  // Before the turn's first samples there is nothing to extrapolate from, so\n  // the gap is plain silence and counted as nothing \u2014 WebRTC likewise only\n  // counts concealment once playout has begun. After that, loop the retained\n  // tail under a decaying gain: a hard zero-fill is a discontinuity mid-word,\n  // which is the click that makes a brief stall sound like breakage.\n  coverGap(out, start) {\n    if (!this.hasPlayed) {\n      out.fill(0, start);\n      return;\n    }\n    if (!this.concealing) {\n      this.concealing = true;\n      this.concealedSilence = false;\n      this.stats.concealmentEvents++;\n    }\n    const total = out.length - start;\n    const len = this.concealLen;\n    let silent = 0;\n    // Second condition: once the fade has decayed to the floor, every sample\n    // the loop would emit is 0 anyway \u2014 bulk-fill instead of running 128\n    // branchy iterations per quantum for the whole tail of a long stall.\n    if (len === 0 || this.concealGain < 0.001) {\n      out.fill(0, start);\n      silent = total;\n    } else {\n      let g = this.concealGain;\n      for (let i = start; i < out.length; i++) {\n        if (g < 0.001) {\n          // The fade has run out: keep counting the gap, but stop looping a\n          // fragment that is now inaudible anyway.\n          out[i] = 0;\n          silent++;\n          continue;\n        }\n        out[i] = this.concealBuf[this.concealPos] * g;\n        this.concealPos = this.concealPos + 1 === len ? 0 : this.concealPos + 1;\n        g *= this.concealDecay;\n      }\n      this.concealGain = g;\n    }\n    this.stats.concealedSamples += total;\n    if (silent > 0) {\n      this.stats.silentConcealedSamples += silent;\n      if (!this.concealedSilence) {\n        this.concealedSilence = true;\n        this.stats.silentConcealmentEvents++;\n      }\n    }\n  }\n\n  // Retain the tail of a rendered quantum as the next gap's concealment\n  // source, and close any episode the real audio just ended.\n  rememberTail(out, n) {\n    const cap = this.concealCapacity;\n    const take = Math.min(n, cap);\n    // Bulk copies (this runs on every cleanly rendered quantum): the tail is\n    // one contiguous source run, landing in at most two ring runs.\n    const tail = out.subarray(n - take, n);\n    const first = Math.min(take, cap - this.concealWrite);\n    this.concealBuf.set(tail.subarray(0, first), this.concealWrite);\n    if (take > first) this.concealBuf.set(tail.subarray(first), 0);\n    this.concealWrite = (this.concealWrite + take) % cap;\n    this.concealLen = Math.min(cap, this.concealLen + take);\n    // Read the loop oldest-first; once the ring is full the write cursor is\n    // the oldest retained sample.\n    this.concealPos = this.concealLen === cap ? this.concealWrite : 0;\n    this.concealing = false;\n    this.concealGain = 1;\n  }\n\n  ingestBytes(uint8) {\n    let bytes = uint8;\n\n    if (this.carry !== null) {\n      const merged = new Uint8Array(1 + bytes.length);\n      merged[0] = this.carry;\n      merged.set(bytes, 1);\n      bytes = merged;\n      this.carry = null;\n    }\n\n    if (bytes.length % 2 !== 0) {\n      this.carry = bytes[bytes.length - 1];\n      bytes = bytes.subarray(0, bytes.length - 1);\n    }\n\n    if (bytes.length === 0) return;\n    const numSamples = bytes.length / 2;\n    const cap = this.capacity;\n    const samples = this.samples;\n    if (this.littleEndian && (bytes.byteOffset & 1) === 0) {\n      // Fast path: 2-byte-aligned LE bytes wrap directly as an Int16Array;\n      // copy wrap-aware in at most two runs with no per-sample DataView call\n      // or modulo. This runs on the realtime audio thread.\n      const int16 = new Int16Array(bytes.buffer, bytes.byteOffset, numSamples);\n      let src = 0;\n      let dst = this.writePos % cap;\n      while (src < numSamples) {\n        const run = Math.min(numSamples - src, cap - dst);\n        for (let j = 0; j < run; j++) {\n          samples[dst + j] = int16[src + j] / 0x8000;\n        }\n        src += run;\n        dst = 0;\n      }\n    } else {\n      // Odd byte offset (or big-endian host): fall back to per-sample reads.\n      const view = new DataView(bytes.buffer, bytes.byteOffset, bytes.length);\n      let dst = this.writePos % cap;\n      for (let i = 0; i < numSamples; i++) {\n        samples[dst] = view.getInt16(i * 2, true) / 0x8000;\n        dst++;\n        if (dst === cap) dst = 0;\n      }\n    }\n    this.writePos += numSamples;\n    // If the producer outran the consumer by more than the buffer holds, drop\n    // the oldest unplayed audio rather than reading samples we've overwritten.\n    if (this.writePos - this.readPos > this.capacity) {\n      this.readPos = this.writePos - this.capacity;\n    }\n  }\n\n  process(inputs, outputs) {\n    // No output wired up yet \u2014 nothing to render this quantum. Throwing here\n    // would permanently kill the processor (the node is persistent per\n    // session), so guard like the capture processor does.\n    if (!outputs[0] || !outputs[0][0]) return true;\n    const out = outputs[0][0];\n    const avail = this.writePos - this.readPos;\n\n    // Report the unplayed backlog to the host (`playback_progress`). This IS\n    // the closed-loop signal: the host otherwise models playback open-loop \u2014\n    // every forwarded chunk assumed to start playing on arrival at exactly\n    // 1.0x \u2014 and cannot see a buffer that has run ahead of the wall clock.\n    // Measured against a client draining at 0.6x, the host declared the line\n    // silent while it still held seconds of the reply, and then opened the\n    // speaking-edge gate over speech the caller had not heard.\n    //\n    // Sent from process() rather than a timer because this is the only place\n    // that sees the buffer on the audio thread; the counter is in QUANTA, so\n    // it costs an integer compare per render and no clock read. Silence is not\n    // reported: an empty buffer is what the host already assumes.\n    this.sinceReportSamples += out.length;\n    if (this.sinceReportSamples >= this.reportIntervalSamples) {\n      this.sinceReportSamples = 0;\n      if (avail > 0) {\n        this.port.postMessage({ event: 'progress', bufferedMs: (avail / sampleRate) * 1000 });\n      }\n    }\n\n    // Filling: wait for the target. 'done' short-circuits it \u2014 what is\n    // buffered is all there will be, so there is nothing left to wait for.\n    if (!this.playing) {\n      if (avail >= this.fillTarget || this.isDone) {\n        this.playing = true;\n      } else {\n        this.coverGap(out, 0);\n        return true;\n      }\n    }\n\n    // Underrun: this quantum cannot be filled and more audio is still coming.\n    // Go back to filling (at the refill target) and cover the gap, leaving\n    // readPos untouched \u2014 the fragment stays buffered and plays intact once\n    // the buffer recovers, instead of being dribbled out a few samples at a\n    // time for the rest of the turn.\n    if (avail < out.length && !this.isDone) {\n      this.playing = false;\n      this.fillTarget = this.refillSamples;\n      this.coverGap(out, 0);\n      return true;\n    }\n\n    if (avail > 0) {\n      const n = Math.min(avail, out.length);\n      // Copy from the ring buffer, splitting across the wrap boundary.\n      const start = this.readPos % this.capacity;\n      const first = Math.min(n, this.capacity - start);\n      out.set(this.samples.subarray(start, start + first), 0);\n      if (n > first) out.set(this.samples.subarray(0, n - first), first);\n      this.readPos += n;\n      // Only reachable with n < out.length on the turn's final partial\n      // quantum (the underrun branch above catches every other case).\n      out.fill(0, n);\n      this.hasPlayed = true;\n      this.rememberTail(out, n);\n      return true;\n    }\n\n    // Drained and done: end the turn. Not reachable mid-turn \u2014 an empty\n    // buffer with audio still coming is the underrun branch above.\n    out.fill(0);\n    if (this.isDone) {\n      this.stopTurn('done');\n    }\n    return true;\n  }\n}\n\nregisterProcessor('playback-processor', PlaybackProcessor);\n";
 declare const _default: string;
 export default _default;

package/dist/worklets/playback-processor.js

@@ -1,4 +1,4 @@
-import { PLAYBACK_BUFFER_SECONDS, PLAYBACK_CONCEAL_FADE_MS, PLAYBACK_CONCEAL_FLOOR, PLAYBACK_JITTER_MS, PLAYBACK_REFILL_MS } from "../types.js";
+import { PLAYBACK_BUFFER_SECONDS, PLAYBACK_CONCEAL_FADE_MS, PLAYBACK_CONCEAL_FLOOR, PLAYBACK_JITTER_MS, PLAYBACK_PROGRESS_INTERVAL_MS, PLAYBACK_REFILL_MS } from "../types.js";
 import { t as workletModuleUrl } from "../_module-url-C_4gRVL0.js";
 //#region worklets/playback-processor.ts
 const PlaybackProcessorWorklet = `
@@ -29,6 +29,11 @@ class PlaybackProcessor extends AudioWorkletProcessor {
     // playing instead of writing past the end and going silent.
     this.capacity = rate * ${PLAYBACK_BUFFER_SECONDS};
     this.samples = new Float32Array(this.capacity);
+    // Cadence for the 'progress' report in process(), counted in samples so
+    // the hot path never reads a clock. Survives resetTurn(): the host clamps
+    // upward only, so a stale count costs at most one extra report.
+    this.reportIntervalSamples = Math.floor((rate * ${PLAYBACK_PROGRESS_INTERVAL_MS}) / 1000);
+    this.sinceReportSamples = 0;
     // Platform endianness probe: the wire format is PCM16 little-endian, so
     // the Int16Array fast path in ingestBytes is only valid on LE hosts
     // (every shipping browser target; the DataView path is the fallback).
@@ -48,6 +53,10 @@ class PlaybackProcessor extends AudioWorkletProcessor {
         this.stopTurn('interrupt');
       } else if (d.event === 'done') {
         this.isDone = true;
+        // Echoed back on this turn's 'stop' so the host can tell WHICH turn
+        // drained: a stop already in flight when a barge-in lands would
+        // otherwise settle the next turn's done() the moment it arrives.
+        this.doneTurn = d.turn ?? null;
       }
     };
   }
@@ -56,6 +65,8 @@ class PlaybackProcessor extends AudioWorkletProcessor {
   // reallocating the sample buffer or re-instantiating the worklet.
   resetTurn() {
     this.isDone = false;
+    // Host turn id carried by this turn's 'done' (null until one arrives).
+    this.doneTurn = null;
     this.playing = false;
     // Whether any real audio has been rendered this turn. Separates a turn's
     // pre-roll (nothing to extrapolate from, and not a defect) from a
@@ -88,10 +99,12 @@ class PlaybackProcessor extends AudioWorkletProcessor {
   // Must NOT return false from process() — a processor that stops is dead
   // for good, forcing a new node (and buffer) per reply.
   // \`reason\` ('interrupt' | 'done') tells the host which turn boundary this
-  // stop belongs to: interrupt-stops are dropped host-side (flush() already
-  // settled that turn), so they can never resolve a later turn's done() early.
+  // stop belongs to — interrupt-stops are dropped host-side, flush() having
+  // already settled that turn — and \`turn\` names WHICH turn drained (the id
+  // this turn's 'done' carried), so a drain-stop still in flight when a
+  // barge-in lands cannot settle the next turn's done() early.
   stopTurn(reason) {
-    this.port.postMessage({ event: 'stop', reason, stats: this.stats });
+    this.port.postMessage({ event: 'stop', reason, turn: this.doneTurn, stats: this.stats });
     this.resetTurn();
   }
 
@@ -228,6 +241,26 @@ class PlaybackProcessor extends AudioWorkletProcessor {
     const out = outputs[0][0];
     const avail = this.writePos - this.readPos;
 
+    // Report the unplayed backlog to the host (\`playback_progress\`). This IS
+    // the closed-loop signal: the host otherwise models playback open-loop —
+    // every forwarded chunk assumed to start playing on arrival at exactly
+    // 1.0x — and cannot see a buffer that has run ahead of the wall clock.
+    // Measured against a client draining at 0.6x, the host declared the line
+    // silent while it still held seconds of the reply, and then opened the
+    // speaking-edge gate over speech the caller had not heard.
+    //
+    // Sent from process() rather than a timer because this is the only place
+    // that sees the buffer on the audio thread; the counter is in QUANTA, so
+    // it costs an integer compare per render and no clock read. Silence is not
+    // reported: an empty buffer is what the host already assumes.
+    this.sinceReportSamples += out.length;
+    if (this.sinceReportSamples >= this.reportIntervalSamples) {
+      this.sinceReportSamples = 0;
+      if (avail > 0) {
+        this.port.postMessage({ event: 'progress', bufferedMs: (avail / sampleRate) * 1000 });
+      }
+    }
+
     // Filling: wait for the target. 'done' short-circuits it — what is
     // buffered is all there will be, so there is nothing left to wait for.
     if (!this.playing) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@alexkroman1/aai-ui",
-  "version": "5.9.0",
+  "version": "5.10.0",
   "type": "module",
   "files": [
     "dist",
@@ -12,6 +12,11 @@
       "types": "./dist/index.d.ts",
       "import": "./dist/index.js"
     },
+    "./client-dir": {
+      "@dev/source": "./client-dir.ts",
+      "types": "./dist/client-dir.d.ts",
+      "import": "./dist/client-dir.js"
+    },
     "./styles.css": "./styles.css",
     "./package.json": "./package.json",
     "./default-client/*": "./dist/default-client/*"
@@ -23,7 +28,7 @@
     "remark-gfm": "^4.0.1",
     "use-stick-to-bottom": "^1.1.6",
     "use-sync-external-store": "^1.6.0",
-    "@alexkroman1/aai": "5.9.0"
+    "@alexkroman1/aai": "5.10.0"
   },
   "peerDependencies": {
     "react": "^19.0.0",
@@ -43,11 +48,11 @@
     "@types/react-dom": "^19.2.3",
     "@types/use-sync-external-store": "^1.5.0",
     "@vitejs/plugin-react": "^6.0.4",
+    "fast-check": "^4.9.0",
     "jsdom": "^29.1.1",
     "react": "^19.2.8",
     "react-dom": "^19.2.8",
     "tsdown": "^0.22.13",
-    "tsx": "^4.23.1",
     "vite": "^8.1.5",
     "vitest": "^4.1.10"
   },
@@ -62,7 +67,7 @@
   "scripts": {
     "test": "vitest run",
     "test:coverage": "vitest run --coverage",
-    "build": "tsdown && tsc -p tsconfig.build.json && tsx build-default-client.ts",
+    "build": "tsdown && tsc -p tsconfig.build.json && node build-default-client.ts",
     "typecheck": "tsc --noEmit",
     "lint": "biome check .",
     "check:publint": "publint",

package/dist/default-client/assets/audio-v_cWeUOV.js

@@ -1 +0,0 @@
-import{a as e,o as t,t as n}from"./index-fLjOiK97.js";function r(e,t,n){if(e!==t)throw Error(`Browser refused the ${n} sample rate: asked for ${t} Hz, got ${e} Hz`)}function i(e){e.then(e=>{for(let t of e.getTracks())t.stop()}).catch(()=>{})}function a(e,t,n){let r=new AudioWorkletNode(e,`capture-processor`,{channelCount:1,channelCountMode:`explicit`}),i=null;return r.port.onmessage=e=>{let r=e.data;r.event===`chunk`&&r.buffer?t(r.buffer):r.event===`silent`?n?.():r.event===`stopped`&&(i?.(),i=null)},{node:r,start(){r.port.postMessage({event:`start`})},stop(){return new Promise(e=>{let t=setTimeout(e,250);i=()=>{clearTimeout(t),e()},r.port.postMessage({event:`stop`})})}}}async function o(o){let{sttSampleRate:s,ttsSampleRate:c,captureWorkletSrc:l,playbackWorkletSrc:u,onMicData:d,onError:f,onPlaybackStats:p,onMicSilent:m}=o,h=new AudioContext({sampleRate:c,latencyHint:`playback`}),g=s===c,_=g?h:new AudioContext({sampleRate:s,latencyHint:`interactive`});async function v(){let e=g?[h]:[h,_];await Promise.all(e.map(e=>e.close().catch(e=>{console.warn(`AudioContext close failed:`,e)})))}let y=navigator.mediaDevices.getUserMedia({audio:{deviceId:{ideal:`default`},...n}}),b;try{[b]=await Promise.all([y,h.resume(),_.resume(),_.audioWorklet.addModule(l),h.audioWorklet.addModule(u)]),r(_.sampleRate,s,`capture`),r(h.sampleRate,c,`playback`)}catch(e){throw i(y),await v(),e}let x=_.createMediaStreamSource(b),S=a(_,d,m);x.connect(S.node),S.node.onprocessorerror=()=>{let e=Error(`Audio capture worklet crashed`);console.error(`[aai-ui]`,e.message),f?.(e)},S.start();let C=null,w=null,T=new AbortController;function E(){if(C)return C;let e=new AudioWorkletNode(h,`playback-processor`);return e.connect(h.destination),e.port.onmessage=e=>{if(e.data.event===`stop`){let t=e.data.stats;if(t&&t.concealedSamples>0&&p?.(t),e.data.reason===`interrupt`)return;w?.(),w=null}},e.onprocessorerror=()=>{let e=Error(`Audio playback worklet crashed`);console.error(`[aai-ui]`,e.message),w?.(),w=null,f?.(e)},C=e,e}let D={enqueue(e){T.signal.aborted||e.byteLength!==0&&E().port.postMessage({event:`write`,buffer:new Uint8Array(e)},[e])},done(){return!C||(C.port.postMessage({event:`done`}),h.state!==`running`)?Promise.resolve():new Promise(n=>{w?.();let r=()=>{clearInterval(i),clearTimeout(a),w===r&&(w=null),n()},i=setInterval(()=>{h.state!==`running`&&r()},t),a=setTimeout(r,e);w=r})},flush(){C&&(w?.(),w=null,C.port.postMessage({event:`interrupt`}))},async close(){if(!T.signal.aborted){T.abort(),await S.stop(),x.disconnect(),S.node.disconnect(),C&&C.disconnect();for(let e of b.getTracks())e.stop();await v()}},async[Symbol.asyncDispose](){await D.close()}};return D}export{o as createVoiceIO};

package/dist/default-client/assets/index-CuPmIAbL.css

@@ -1,2 +0,0 @@
-/*! tailwindcss v4.3.3 | MIT License | https://tailwindcss.com */
-@layer properties{@supports (((-webkit-hyphens:none)) and (not (margin-trim:inline))) or ((-moz-orient:inline) and (not (color:rgb(from red r g b)))){*,:before,:after,::backdrop{--tw-border-style:solid;--tw-leading:initial;--tw-font-weight:initial;--tw-tracking:initial;--tw-shadow:0 0 #0000;--tw-shadow-color:initial;--tw-shadow-alpha:100%;--tw-inset-shadow:0 0 #0000;--tw-inset-shadow-color:initial;--tw-inset-shadow-alpha:100%;--tw-ring-color:initial;--tw-ring-shadow:0 0 #0000;--tw-inset-ring-color:initial;--tw-inset-ring-shadow:0 0 #0000;--tw-ring-inset:initial;--tw-ring-offset-width:0px;--tw-ring-offset-color:#fff;--tw-ring-offset-shadow:0 0 #0000;--tw-blur:initial;--tw-brightness:initial;--tw-contrast:initial;--tw-grayscale:initial;--tw-hue-rotate:initial;--tw-invert:initial;--tw-opacity:initial;--tw-saturate:initial;--tw-sepia:initial;--tw-drop-shadow:initial;--tw-drop-shadow-color:initial;--tw-drop-shadow-alpha:100%;--tw-drop-shadow-size:initial;--tw-duration:initial;--tw-ease:initial}}}@layer theme{:root,:host{--font-sans:-apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, "Helvetica Neue", "Noto Sans", Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol", "Noto Color Emoji";--font-mono:ui-monospace, SFMono-Regular, Menlo, Monaco, Consolas, "Liberation Mono", "Courier New", monospace;--spacing:.25rem;--text-xs:.75rem;--text-xs--line-height:calc(1 / .75);--text-sm:.875rem;--text-sm--line-height:calc(1.25 / .875);--font-weight-normal:400;--font-weight-medium:500;--font-weight-semibold:600;--tracking-wide:.025em;--radius-sm:.25rem;--radius-md:.375rem;--radius-lg:.5rem;--radius-xl:.75rem;--ease-in-out:cubic-bezier(.4, 0, .2, 1);--default-transition-duration:.15s;--default-transition-timing-function:cubic-bezier(.4, 0, .2, 1);--default-font-family:var(--font-sans);--default-mono-font-family:var(--font-mono);--font-aai:"Monument Grotesk", "ABC Monument Grotesk", ui-sans-serif, system-ui, -apple-system, sans-serif;--font-aai-serif:"Source Serif 4", "Source Serif Pro", Charter, "Iowan Old Style", Georgia, serif;--font-aai-mono:"JetBrains Mono", ui-monospace, SFMono-Regular, Menlo, monospace;--radius-aai:4px}}@layer base{*,:after,:before,::backdrop{box-sizing:border-box;border:0 solid;margin:0;padding:0}::file-selector-button{box-sizing:border-box;border:0 solid;margin:0;padding:0}html,:host{-webkit-text-size-adjust:100%;tab-size:4;line-height:1.5;font-family:var(--default-font-family,-apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, "Helvetica Neue", "Noto Sans", Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol", "Noto Color Emoji");font-feature-settings:var(--default-font-feature-settings,normal);font-variation-settings:var(--default-font-variation-settings,normal);-webkit-tap-highlight-color:transparent}hr{height:0;color:inherit;border-top-width:1px}abbr:where([title]){-webkit-text-decoration:underline dotted;text-decoration:underline dotted}h1,h2,h3,h4,h5,h6{font-size:inherit;font-weight:inherit}a{color:inherit;-webkit-text-decoration:inherit;-webkit-text-decoration:inherit;-webkit-text-decoration:inherit;-webkit-text-decoration:inherit;text-decoration:inherit}b,strong{font-weight:bolder}code,kbd,samp,pre{font-family:var(--default-mono-font-family,ui-monospace, SFMono-Regular, Menlo, Monaco, Consolas, "Liberation Mono", "Courier New", monospace);font-feature-settings:var(--default-mono-font-feature-settings,normal);font-variation-settings:var(--default-mono-font-variation-settings,normal);font-size:1em}small{font-size:80%}sub,sup{vertical-align:baseline;font-size:75%;line-height:0;position:relative}sub{bottom:-.25em}sup{top:-.5em}table{text-indent:0;border-color:inherit;border-collapse:collapse}:-moz-focusring:where(:not(iframe)){outline:auto}progress{vertical-align:baseline}summary{display:list-item}ol,ul,menu{list-style:none}img,svg,video,canvas,audio,iframe,embed,object{vertical-align:middle;display:block}img,video{max-width:100%;height:auto}button,input,select,optgroup,textarea{font:inherit;font-feature-settings:inherit;font-variation-settings:inherit;letter-spacing:inherit;color:inherit;opacity:1;background-color:#0000;border-radius:0}::file-selector-button{font:inherit;font-feature-settings:inherit;font-variation-settings:inherit;letter-spacing:inherit;color:inherit;opacity:1;background-color:#0000;border-radius:0}:where(select:is([multiple],[size])) optgroup{font-weight:bolder}:where(select:is([multiple],[size])) optgroup option{padding-inline-start:20px}::file-selector-button{margin-inline-end:4px}::placeholder{opacity:1}@supports (not ((-webkit-appearance:-apple-pay-button))) or (contain-intrinsic-size:1px){::placeholder{color:currentColor}@supports (color:color-mix(in lab, red, red)){::placeholder{color:color-mix(in oklab, currentcolor 50%, transparent)}}}textarea{resize:vertical}::-webkit-search-decoration{-webkit-appearance:none}::-webkit-date-and-time-value{min-height:1lh;text-align:inherit}::-webkit-datetime-edit{display:inline-flex}::-webkit-datetime-edit-fields-wrapper{padding:0}::-webkit-datetime-edit{padding-block:0}::-webkit-datetime-edit-year-field{padding-block:0}::-webkit-datetime-edit-month-field{padding-block:0}::-webkit-datetime-edit-day-field{padding-block:0}::-webkit-datetime-edit-hour-field{padding-block:0}::-webkit-datetime-edit-minute-field{padding-block:0}::-webkit-datetime-edit-second-field{padding-block:0}::-webkit-datetime-edit-millisecond-field{padding-block:0}::-webkit-datetime-edit-meridiem-field{padding-block:0}::-webkit-calendar-picker-indicator{line-height:1}:-moz-ui-invalid{box-shadow:none}button,input:where([type=button],[type=reset],[type=submit]){appearance:button}::file-selector-button{appearance:button}::-webkit-inner-spin-button{height:auto}::-webkit-outer-spin-button{height:auto}[hidden]:where(:not([hidden=until-found])){display:none!important}html,body{margin:0;padding:0}}@layer components;@layer utilities{.collapse{visibility:collapse}.invisible{visibility:hidden}.visible{visibility:visible}.absolute{position:absolute}.relative{position:relative}.static{position:static}.container{width:100%}@media (width>=40rem){.container{max-width:40rem}}@media (width>=48rem){.container{max-width:48rem}}@media (width>=64rem){.container{max-width:64rem}}@media (width>=80rem){.container{max-width:80rem}}@media (width>=96rem){.container{max-width:96rem}}.m-0{margin:0}.mx-auto{margin-inline:auto}.my-0\.5{margin-block:calc(var(--spacing) * .5)}.my-1\.5{margin-block:calc(var(--spacing) * 1.5)}.my-2\.5{margin-block:calc(var(--spacing) * 2.5)}.mt-2{margin-top:calc(var(--spacing) * 2)}.mt-3{margin-top:calc(var(--spacing) * 3)}.mb-1\.5{margin-bottom:calc(var(--spacing) * 1.5)}.ml-auto{margin-left:auto}.box-border{box-sizing:border-box}.block{display:block}.flex{display:flex}.grid{display:grid}.inline{display:inline}.inline-flex{display:inline-flex}.table{display:table}.h-1\.5{height:calc(var(--spacing) * 1.5)}.h-4{height:calc(var(--spacing) * 4)}.h-9{height:calc(var(--spacing) * 9)}.h-11{height:calc(var(--spacing) * 11)}.h-\[7px\]{height:7px}.h-screen{height:100vh}.max-h-64{max-height:calc(var(--spacing) * 64)}.min-h-0{min-height:0}.min-h-5{min-height:calc(var(--spacing) * 5)}.w-1\.5{width:calc(var(--spacing) * 1.5)}.w-4{width:calc(var(--spacing) * 4)}.w-\[7px\]{width:7px}.w-fit{width:fit-content}.w-full{width:100%}.max-w-75{max-width:calc(var(--spacing) * 75)}.max-w-105{max-width:calc(var(--spacing) * 105)}.max-w-190{max-width:calc(var(--spacing) * 190)}.max-w-\[60\%\]{max-width:60%}.max-w-\[82\%\]{max-width:82%}.max-w-\[min\(78\%\,64ch\)\]{max-width:min(78%,64ch)}.min-w-0{min-width:0}.flex-1{flex:1}.shrink-0{flex-shrink:0}.border-collapse{border-collapse:collapse}.rotate-90{rotate:90deg}.cursor-pointer{cursor:pointer}.resize{resize:both}.\[scrollbar-width\:none\]{scrollbar-width:none}.list-decimal{list-style-type:decimal}.list-disc{list-style-type:disc}.appearance-none{appearance:none}.flex-col{flex-direction:column}.items-center{align-items:center}.items-end{align-items:flex-end}.justify-between{justify-content:space-between}.justify-center{justify-content:center}.justify-end{justify-content:flex-end}.gap-1{gap:var(--spacing)}.gap-1\.5{gap:calc(var(--spacing) * 1.5)}.gap-2{gap:calc(var(--spacing) * 2)}.gap-2\.5{gap:calc(var(--spacing) * 2.5)}.gap-3{gap:calc(var(--spacing) * 3)}.gap-4{gap:calc(var(--spacing) * 4)}.gap-5{gap:calc(var(--spacing) * 5)}.truncate{text-overflow:ellipsis;white-space:nowrap;overflow:hidden}.overflow-auto{overflow:auto}.overflow-hidden{overflow:hidden}.overflow-x-auto{overflow-x:auto}.overflow-y-auto{overflow-y:auto}.rounded-aai{border-radius:var(--radius-aai)}.rounded-full{border-radius:3.40282e38px}.rounded-lg{border-radius:var(--radius-lg)}.rounded-md{border-radius:var(--radius-md)}.rounded-sm{border-radius:var(--radius-sm)}.rounded-xl{border-radius:var(--radius-xl)}.border{border-style:var(--tw-border-style);border-width:1px}.border-t{border-top-style:var(--tw-border-style);border-top-width:1px}.border-b{border-bottom-style:var(--tw-border-style);border-bottom-width:1px}.border-l-2{border-left-style:var(--tw-border-style);border-left-width:2px}.border-none{--tw-border-style:none;border-style:none}.bg-transparent{background-color:#0000}.p-2\.5{padding:calc(var(--spacing) * 2.5)}.p-7{padding:calc(var(--spacing) * 7)}.px-1{padding-inline:var(--spacing)}.px-2{padding-inline:calc(var(--spacing) * 2)}.px-3{padding-inline:calc(var(--spacing) * 3)}.px-3\.5{padding-inline:calc(var(--spacing) * 3.5)}.px-5{padding-inline:calc(var(--spacing) * 5)}.px-6{padding-inline:calc(var(--spacing) * 6)}.px-7{padding-inline:calc(var(--spacing) * 7)}.px-10{padding-inline:calc(var(--spacing) * 10)}.py-0\.5{padding-block:calc(var(--spacing) * .5)}.py-1{padding-block:var(--spacing)}.py-2{padding-block:calc(var(--spacing) * 2)}.py-2\.5{padding-block:calc(var(--spacing) * 2.5)}.py-3{padding-block:calc(var(--spacing) * 3)}.py-8{padding-block:calc(var(--spacing) * 8)}.py-12{padding-block:calc(var(--spacing) * 12)}.pl-3{padding-left:calc(var(--spacing) * 3)}.pl-5{padding-left:calc(var(--spacing) * 5)}.text-center{text-align:center}.text-left{text-align:left}.font-aai{font-family:var(--font-aai)}.font-aai-mono{font-family:var(--font-aai-mono)}.font-aai-serif{font-family:var(--font-aai-serif)}.text-sm{font-size:var(--text-sm);line-height:var(--tw-leading,var(--text-sm--line-height))}.text-xs{font-size:var(--text-xs);line-height:var(--tw-leading,var(--text-xs--line-height))}.text-\[9px\]{font-size:9px}.text-\[10px\]{font-size:10px}.text-\[11px\]{font-size:11px}.text-\[12\.5px\]{font-size:12.5px}.text-\[12px\]{font-size:12px}.text-\[13px\]{font-size:13px}.text-\[14px\]{font-size:14px}.text-\[15px\]{font-size:15px}.text-\[16px\]{font-size:16px}.text-\[17px\]{font-size:17px}.text-\[22px\]{font-size:22px}.text-\[32px\]{font-size:32px}.leading-4{--tw-leading:calc(var(--spacing) * 4);line-height:calc(var(--spacing) * 4)}.leading-\[1\.2\]{--tw-leading:1.2;line-height:1.2}.leading-\[1\.15\]{--tw-leading:1.15;line-height:1.15}.leading-\[22px\]{--tw-leading:22px;line-height:22px}.leading-\[23px\]{--tw-leading:23px;line-height:23px}.leading-\[130\%\]{--tw-leading:130%;line-height:130%}.leading-none{--tw-leading:1;line-height:1}.font-medium{--tw-font-weight:var(--font-weight-medium);font-weight:var(--font-weight-medium)}.font-normal{--tw-font-weight:var(--font-weight-normal);font-weight:var(--font-weight-normal)}.font-semibold{--tw-font-weight:var(--font-weight-semibold);font-weight:var(--font-weight-semibold)}.tracking-\[-0\.2px\]{--tw-tracking:-.2px;letter-spacing:-.2px}.tracking-\[1\.2px\]{--tw-tracking:1.2px;letter-spacing:1.2px}.tracking-\[1\.4px\]{--tw-tracking:1.4px;letter-spacing:1.4px}.tracking-wide{--tw-tracking:var(--tracking-wide);letter-spacing:var(--tracking-wide)}.text-balance{text-wrap:balance}.wrap-break-word{overflow-wrap:break-word}.whitespace-nowrap{white-space:nowrap}.whitespace-pre-wrap{white-space:pre-wrap}.uppercase{text-transform:uppercase}.underline{text-decoration-line:underline}.underline-offset-2{text-underline-offset:2px}.ring{--tw-ring-shadow:var(--tw-ring-inset,) 0 0 0 calc(1px + var(--tw-ring-offset-width)) var(--tw-ring-color,currentcolor);box-shadow:var(--tw-inset-shadow), var(--tw-inset-ring-shadow), var(--tw-ring-offset-shadow), var(--tw-ring-shadow), var(--tw-shadow)}.filter{filter:var(--tw-blur,) var(--tw-brightness,) var(--tw-contrast,) var(--tw-grayscale,) var(--tw-hue-rotate,) var(--tw-invert,) var(--tw-saturate,) var(--tw-sepia,) var(--tw-drop-shadow,)}.transition{transition-property:color,background-color,border-color,outline-color,text-decoration-color,fill,stroke,--tw-gradient-from,--tw-gradient-via,--tw-gradient-to,opacity,box-shadow,transform,translate,scale,rotate,filter,-webkit-backdrop-filter,backdrop-filter,display,content-visibility,overlay,pointer-events;transition-timing-function:var(--tw-ease,var(--default-transition-timing-function));transition-duration:var(--tw-duration,var(--default-transition-duration))}.transition-colors{transition-property:color,background-color,border-color,outline-color,text-decoration-color,fill,stroke,--tw-gradient-from,--tw-gradient-via,--tw-gradient-to;transition-timing-function:var(--tw-ease,var(--default-transition-timing-function));transition-duration:var(--tw-duration,var(--default-transition-duration))}.transition-transform{transition-property:transform,translate,scale,rotate;transition-timing-function:var(--tw-ease,var(--default-transition-timing-function));transition-duration:var(--tw-duration,var(--default-transition-duration))}.duration-150{--tw-duration:.15s;transition-duration:.15s}.ease-in-out{--tw-ease:var(--ease-in-out);transition-timing-function:var(--ease-in-out)}.outline-none{--tw-outline-style:none;outline-style:none}.select-none{-webkit-user-select:none;user-select:none}.first\:mt-0:first-child{margin-top:0}.last\:mb-0:last-child{margin-bottom:0}.disabled\:cursor-not-allowed:disabled{cursor:not-allowed}.disabled\:opacity-50:disabled{opacity:.5}@media (width>=40rem){.sm\:px-16{padding-inline:calc(var(--spacing) * 16)}.sm\:py-14{padding-block:calc(var(--spacing) * 14)}}}@keyframes aai-pulse{0%,to{opacity:1;transform:scale(1)}50%{opacity:.45;transform:scale(.82)}}@keyframes aai-bounce{0%,80%,to{opacity:.3;transform:scale(.8)}40%{opacity:1;transform:scale(1)}}@keyframes aai-shimmer{0%{background-position:-200% 0}to{background-position:200% 0}}.tool-shimmer{-webkit-text-fill-color:transparent;background:linear-gradient(90deg,currentColor 25%,#0000 50%,currentColor 75%) 0 0/200% 100%;-webkit-background-clip:text;background-clip:text;animation:2s infinite aai-shimmer}@property --tw-border-style{syntax:"*";inherits:false;initial-value:solid}@property --tw-leading{syntax:"*";inherits:false}@property --tw-font-weight{syntax:"*";inherits:false}@property --tw-tracking{syntax:"*";inherits:false}@property --tw-shadow{syntax:"*";inherits:false;initial-value:0 0 #0000}@property --tw-shadow-color{syntax:"*";inherits:false}@property --tw-shadow-alpha{syntax:"<percentage>";inherits:false;initial-value:100%}@property --tw-inset-shadow{syntax:"*";inherits:false;initial-value:0 0 #0000}@property --tw-inset-shadow-color{syntax:"*";inherits:false}@property --tw-inset-shadow-alpha{syntax:"<percentage>";inherits:false;initial-value:100%}@property --tw-ring-color{syntax:"*";inherits:false}@property --tw-ring-shadow{syntax:"*";inherits:false;initial-value:0 0 #0000}@property --tw-inset-ring-color{syntax:"*";inherits:false}@property --tw-inset-ring-shadow{syntax:"*";inherits:false;initial-value:0 0 #0000}@property --tw-ring-inset{syntax:"*";inherits:false}@property --tw-ring-offset-width{syntax:"<length>";inherits:false;initial-value:0}@property --tw-ring-offset-color{syntax:"*";inherits:false;initial-value:#fff}@property --tw-ring-offset-shadow{syntax:"*";inherits:false;initial-value:0 0 #0000}@property --tw-blur{syntax:"*";inherits:false}@property --tw-brightness{syntax:"*";inherits:false}@property --tw-contrast{syntax:"*";inherits:false}@property --tw-grayscale{syntax:"*";inherits:false}@property --tw-hue-rotate{syntax:"*";inherits:false}@property --tw-invert{syntax:"*";inherits:false}@property --tw-opacity{syntax:"*";inherits:false}@property --tw-saturate{syntax:"*";inherits:false}@property --tw-sepia{syntax:"*";inherits:false}@property --tw-drop-shadow{syntax:"*";inherits:false}@property --tw-drop-shadow-color{syntax:"*";inherits:false}@property --tw-drop-shadow-alpha{syntax:"<percentage>";inherits:false;initial-value:100%}@property --tw-drop-shadow-size{syntax:"*";inherits:false}@property --tw-duration{syntax:"*";inherits:false}@property --tw-ease{syntax:"*";inherits:false}