realtime-avatar 0.31.0 → 0.32.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +6 -0
- package/dist/{chunk-TJM53JXR.js → chunk-NDCZABOK.js} +1 -1
- package/dist/{chunk-LRGQKZFB.js → chunk-NKS5ZIWN.js} +1 -1
- package/dist/express.d.ts +2 -2
- package/dist/express.js +2 -2
- package/dist/hono.d.ts +2 -2
- package/dist/hono.js +2 -2
- package/dist/index.d.ts +2 -2
- package/dist/index.js +1 -1
- package/dist/nextjs.d.ts +2 -2
- package/dist/nextjs.js +2 -2
- package/dist/react-native.d.ts +9 -2
- package/dist/react-native.js +49 -2
- package/dist/server.d.ts +1 -1
- package/dist/server.js +1 -1
- package/dist/tanstack-start.d.ts +2 -2
- package/dist/tanstack-start.js +2 -2
- package/dist/{types-DOFXGl7G.d.ts → types-BSSqsu_Z.d.ts} +31 -1
- package/dist/{types-Du6T6XQB.d.ts → types-su27kEw0.d.ts} +1 -1
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -53,6 +53,12 @@ lifecycle; the hook is not a shared audio-session lease. Camera publication stil
|
|
|
53
53
|
server grant's permission and the caller's opt-in. Room callbacks/options continue to use LiveKit's
|
|
54
54
|
public types and implementation; no separate connection state machine is introduced.
|
|
55
55
|
|
|
56
|
+
**On iOS, the room also clears an input mute an earlier call left on the app.** A server-closed room
|
|
57
|
+
can leave iOS 26's per-app input mute set, so later calls publish an unmuted track of silence. Each
|
|
58
|
+
time the room's own microphone goes live, `RealtimeAvatarLiveKitRoom` round-trips the audio engine's
|
|
59
|
+
mute once so the engine writes "unmuted" itself (iOS with the Voice Processing mute mode only; never
|
|
60
|
+
on Android). A room you compose yourself needs `useClearStaleInputMute(room)` inside it.
|
|
61
|
+
|
|
56
62
|
**This is a native build upgrade, not an OTA-only change.** Rebuild iOS and Android binaries with
|
|
57
63
|
RN3/WebRTC 144.2.0 before shipping JavaScript that requires them. Retain the previous SDK group for
|
|
58
64
|
installed binaries that still use RN2, and isolate their OTA runtime/channel appropriately.
|
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
import { sessionModeSchema, liveKitSessionReleaseReasonSchema, RealtimeAvatar, ROUTE_TIMEOUT_MS, isQueued, RealtimeAvatarHttpError, RealtimeAvatarError } from './chunk-
|
|
1
|
+
import { sessionModeSchema, liveKitSessionReleaseReasonSchema, RealtimeAvatar, ROUTE_TIMEOUT_MS, isQueued, RealtimeAvatarHttpError, RealtimeAvatarError } from './chunk-NKS5ZIWN.js';
|
|
2
2
|
import { z } from 'zod';
|
|
3
3
|
|
|
4
4
|
// ../http-client/src/generated/error-semantics.ts
|
|
@@ -919,7 +919,7 @@ function toSessionsPage(page) {
|
|
|
919
919
|
// ../http-client/src/client.ts
|
|
920
920
|
var DEFAULT_BASE_URL = "https://realtimeavatar.ai/api/v1";
|
|
921
921
|
var RENDER_MISMATCH = "Render response did not match the SDK contract";
|
|
922
|
-
var SDK_VERSION = "0.
|
|
922
|
+
var SDK_VERSION = "0.32.0";
|
|
923
923
|
var RealtimeAvatar = class {
|
|
924
924
|
#apiKey;
|
|
925
925
|
#baseUrl;
|
package/dist/express.d.ts
CHANGED
package/dist/express.js
CHANGED
package/dist/hono.d.ts
CHANGED
package/dist/hono.js
CHANGED
package/dist/index.d.ts
CHANGED
|
@@ -1,5 +1,5 @@
|
|
|
1
|
-
import { a as CallPolicy, C as CallMode, S as StartCallResult, E as EndCallOptions, A as Avatar, b as AvatarUpdate, c as AvatarSourceSwap, d as ClipLibraryDeclaration, e as ClipLibraryUpdate, L as LoopRedirect, f as ClipLibrary, g as AssetKind, h as Asset, i as ListSessionsOptions, U as UsageSessionPage, j as UsageSession, k as ListRecordingsQuery, l as ListRecordingsResponse, R as RecordingArtifact, m as RecordingAccessResponse, n as CreateRenderInput, o as CreditBalance, T as TranscriptPayload, V as VoiceWebhookRequest } from './types-
|
|
2
|
-
export { p as AvatarClip, q as CallConnection, r as CallQueued, s as ClipDeclaration, t as ClipLibraryPlan, u as ClipSource, v as ContextMessage, w as EndCallReason, x as ExternalLlm, y as RecordingArtifactStatus, z as RecordingMode, B as RenderShot, D as VideoPolicy, F as VoiceWebhook, G as VoiceWebhookErrorCode, H as isQueued, I as listRecordingsResponseSchema, J as recordingAccessResponseSchema, K as recordingArtifactSchema, M as recordingModeSchema } from './types-
|
|
1
|
+
import { a as CallPolicy, C as CallMode, S as StartCallResult, E as EndCallOptions, A as Avatar, b as AvatarUpdate, c as AvatarSourceSwap, d as ClipLibraryDeclaration, e as ClipLibraryUpdate, L as LoopRedirect, f as ClipLibrary, g as AssetKind, h as Asset, i as ListSessionsOptions, U as UsageSessionPage, j as UsageSession, k as ListRecordingsQuery, l as ListRecordingsResponse, R as RecordingArtifact, m as RecordingAccessResponse, n as CreateRenderInput, o as CreditBalance, T as TranscriptPayload, V as VoiceWebhookRequest } from './types-BSSqsu_Z.js';
|
|
2
|
+
export { p as AvatarClip, q as CallConnection, r as CallQueued, s as ClipDeclaration, t as ClipLibraryPlan, u as ClipSource, v as ContextMessage, w as EndCallReason, x as ExternalLlm, y as RecordingArtifactStatus, z as RecordingMode, B as RenderShot, D as VideoPolicy, F as VoiceWebhook, G as VoiceWebhookErrorCode, H as isQueued, I as listRecordingsResponseSchema, J as recordingAccessResponseSchema, K as recordingArtifactSchema, M as recordingModeSchema } from './types-BSSqsu_Z.js';
|
|
3
3
|
import { ConnectionHistoryResponse } from './connection-history.js';
|
|
4
4
|
import { z } from 'zod';
|
|
5
5
|
|
package/dist/index.js
CHANGED
|
@@ -1,2 +1,2 @@
|
|
|
1
|
-
export { DEFAULT_FISH_TTS_MODEL, RealtimeAvatar, RealtimeAvatarError, RealtimeAvatarHttpError, RealtimeAvatarWebhookError, clipLibraryDeclarationSchema, fishTtsModelSchema, isQueued, listRecordingsResponseSchema, listRendersResponseSchema, recordingAccessResponseSchema, recordingArtifactSchema, recordingModeSchema, renderAccessResponseSchema, renderSchema, verifyTranscript, verifyVoiceWebhook } from './chunk-
|
|
1
|
+
export { DEFAULT_FISH_TTS_MODEL, RealtimeAvatar, RealtimeAvatarError, RealtimeAvatarHttpError, RealtimeAvatarWebhookError, clipLibraryDeclarationSchema, fishTtsModelSchema, isQueued, listRecordingsResponseSchema, listRendersResponseSchema, recordingAccessResponseSchema, recordingArtifactSchema, recordingModeSchema, renderAccessResponseSchema, renderSchema, verifyTranscript, verifyVoiceWebhook } from './chunk-NKS5ZIWN.js';
|
|
2
2
|
import './chunk-MCIXU6XP.js';
|
package/dist/nextjs.d.ts
CHANGED
package/dist/nextjs.js
CHANGED
package/dist/react-native.d.ts
CHANGED
|
@@ -2,11 +2,12 @@ import { LiveKitRoomProps } from '@livekit/react-native';
|
|
|
2
2
|
export { AudioSession, VideoTrack, VideoTrackProps, registerGlobals } from '@livekit/react-native';
|
|
3
3
|
import { L as LiveKitSessionGrant, A as AvatarVideoFit } from './proxy-client-BlaFxR6S.js';
|
|
4
4
|
export { a as AdaptivePlayoutController, b as AdaptivePlayoutDecision, c as AdaptivePlayoutOptions, d as AdaptivePlayoutSample, e as ApproachingEndEvent, f as ApproachingEndReason, g as AvatarConnectionDetails, h as AvatarSessionClient, B as BehaviorSnapshot, C as CallTranscript, i as CallTranscriptSegment, j as CartesiaTtsModel, k as ClipResult, l as ClosingTurnResult, m as CreditsLowEvent, D as DEFAULT_APPROACHING_END_LEAD_SECONDS, n as DEFAULT_CREDITS_LOW_LEAD_SECONDS, o as DEFAULT_FISH_TTS_MODEL, p as DEFAULT_GOVERNOR_CONFIG, q as DEFAULT_GRACE_CEILING_SECONDS, r as DEFAULT_GRACE_WINDOW_LEAD_SECONDS, s as DEFAULT_IDLE_SECONDS, t as DEFAULT_IDLE_WARN_LEAD_SECONDS, u as DEFAULT_READY_TIMEOUT_SECONDS, v as DEFAULT_TURN_TIMEOUT_SECONDS, w as DeclaredInputSource, E as EndReason, x as EndedEvent, y as ExtendResult, F as FishTtsModel, z as FreezeReadingFn, G as Governor, H as GovernorAction, I as GovernorConfig, J as GovernorSignal, K as GovernorState, M as GraceWindowClosedEvent, N as GraceWindowOpenEvent, O as GraceWindowState, P as IdleWarningEvent, Q as InboundRtpCursor, R as InboundRtpReading, S as KnownBehaviorState, T as LLMProvider, U as LLMSelection, V as LiveKitAvatarGrantState, W as LiveKitAvatarGrantStatus, X as LiveKitCapacityState, Y as LiveKitConnectionStatus, Z as LiveKitSessionRequest, _ as LiveKitSessionStartResult, $ as MAX_RECONNECT_ATTEMPTS, a0 as MAX_SESSION_INSTRUCTIONS_CHARS, a1 as ProxyClientOptions, a2 as QualityCap, a3 as RECONNECT_BACKOFF_MS, a4 as RealtimeAvatarRequestOptions, a5 as RealtimeSessionApi, a6 as RealtimeSessionMedia, a7 as RealtimeSessionRoomSinks, a8 as ReconnectPolicy, a9 as ReconnectingEvent, aa as RecoveryState, ab as RetryStep, ac as SendTurnOptions, ad as SessionBehavior, ae as SessionClip, af as SessionClocks, ag as SessionEndReason, ah as SessionLifecycleApi, ai as SessionLifecyclePhase, aj as SessionLifecyclePhaseKind, ak as SessionLifecycleRoomBridge, al as SessionLifecycleRoomBridgeProps, am as SessionReadinessFailure, an as SurfaceLayers, ao as TranscriptSender, ap as TranscriptSenderOptions, aq as TurnState, ar as TurnTimeoutEvent, as as UseAvatarQualityGovernorInput, at as UseLiveKitAvatarGrantInput, au as UseRealtimeSessionInput, av as UseSessionLifecycleInput, aw as VoiceError, ax as VoiceErrorCode, ay as VoiceSpec, az as VoiceSpecInput, aA as capacityErrorFromBusy, aB as capacityStateFromGrant, aC as createProxyClient, aD as fishTtsModelSchema, aE as isNativeLiveTrackSubscribed, aF as knownBehaviorStates, aG as mapTurnState, aH as readInboundRtp, aI as sessionBehaviorSchema, aJ as sessionClipSchema, aK as splitCallTranscript, aL as useAvatarAdaptivePlayoutDelay, aM as useAvatarCamera, aN as useAvatarPlayoutDelay, aO as useAvatarQualityGovernor, aP as useCallTranscript, aQ as useLiveKitAvatarGrant, aR as useLiveTrackProducing, aS as useMicLease, aT as useRealtimeSession, aU as useReleaseMicLeaseOnTrackEnded, aV as useSessionLifecycle } from './proxy-client-BlaFxR6S.js';
|
|
5
|
+
import { Room } from 'livekit-client';
|
|
6
|
+
export { DisconnectReason, RemoteAudioTrack, RemoteTrack, RemoteVideoTrack, Room, RoomEvent, SendTextOptions, Track } from 'livekit-client';
|
|
5
7
|
import { ReactNode, ReactElement } from 'react';
|
|
6
8
|
import { StyleProp, ViewStyle } from 'react-native';
|
|
7
9
|
export { useChat, useConnectionState, useLocalParticipant, useRoomContext, useTrackToggle, useTranscriptions, useVoiceAssistant } from '@livekit/components-react';
|
|
8
10
|
export { D as DEFAULT_AVATAR_PLAYOUT_DELAY_SECONDS } from './playout-delay-CtRZdtsj.js';
|
|
9
|
-
export { DisconnectReason, RemoteAudioTrack, RemoteTrack, RemoteVideoTrack, Room, RoomEvent, SendTextOptions, Track } from 'livekit-client';
|
|
10
11
|
import 'zod';
|
|
11
12
|
|
|
12
13
|
type RealtimeAvatarLiveKitRoomProps = Omit<LiveKitRoomProps, "serverUrl" | "token" | "connect" | "audio" | "video" | "options"> & {
|
|
@@ -32,6 +33,12 @@ type RealtimeAvatarLiveKitRoomProps = Omit<LiveKitRoomProps, "serverUrl" | "toke
|
|
|
32
33
|
* LiveKitRoom props) can still one-line the session lifecycle.
|
|
33
34
|
*/
|
|
34
35
|
declare function useRealtimeAvatarAudioSession(enabled: boolean): void;
|
|
36
|
+
/**
|
|
37
|
+
* Run `clearStaleInputMute` each time `room`'s local microphone goes live: when it is published
|
|
38
|
+
* unmuted and when it is unmuted later, for as long as that publication stays published and
|
|
39
|
+
* unmuted and this hook stays mounted.
|
|
40
|
+
*/
|
|
41
|
+
declare function useClearStaleInputMute(room: Room | undefined): void;
|
|
35
42
|
/** Thin typed bridge from a Realtime Avatar grant into the native LiveKitRoom. */
|
|
36
43
|
declare function RealtimeAvatarLiveKitRoom(props: RealtimeAvatarLiveKitRoomProps): ReactElement;
|
|
37
44
|
|
|
@@ -126,4 +133,4 @@ type AvatarVideoSurfaceProps = {
|
|
|
126
133
|
*/
|
|
127
134
|
declare function AvatarVideoSurface(props: AvatarVideoSurfaceProps): ReactElement;
|
|
128
135
|
|
|
129
|
-
export { AvatarVideoFit, AvatarVideoSurface, type AvatarVideoSurfaceProps, type IdleVideoRender, RealtimeAvatarLiveKitRoom, type RealtimeAvatarLiveKitRoomProps, useRealtimeAvatarAudioSession };
|
|
136
|
+
export { AvatarVideoFit, AvatarVideoSurface, type AvatarVideoSurfaceProps, type IdleVideoRender, RealtimeAvatarLiveKitRoom, type RealtimeAvatarLiveKitRoomProps, useClearStaleInputMute, useRealtimeAvatarAudioSession };
|
package/dist/react-native.js
CHANGED
|
@@ -3,8 +3,9 @@ export { AudioSession, VideoTrack, registerGlobals } from '@livekit/react-native
|
|
|
3
3
|
import { useState, useRef, useEffect, useCallback, createElement, useMemo } from 'react';
|
|
4
4
|
import { useLocalParticipant, useConnectionState, useTranscriptions, useMaybeRoomContext, useVoiceAssistant, isTrackReference, useRoomContext, useChat, useAudioPlayback } from '@livekit/components-react';
|
|
5
5
|
export { useChat, useConnectionState, useLocalParticipant, useRoomContext, useTrackToggle, useTranscriptions, useVoiceAssistant } from '@livekit/components-react';
|
|
6
|
-
import { ConnectionState, isRemoteTrack, VideoQuality, ConnectionQuality,
|
|
6
|
+
import { ConnectionState, RoomEvent, Track, isRemoteTrack, VideoQuality, ConnectionQuality, TrackEvent, ConnectionError, ConnectionErrorReason, ParticipantEvent, DisconnectReason, createLocalVideoTrack, Room } from 'livekit-client';
|
|
7
7
|
export { DisconnectReason, Room, RoomEvent, Track } from 'livekit-client';
|
|
8
|
+
import { AudioDeviceModule, AudioEngineMuteMode } from '@livekit/react-native-webrtc';
|
|
8
9
|
import { Platform, StyleSheet, Animated, Image, View, Text } from 'react-native';
|
|
9
10
|
import { z } from 'zod';
|
|
10
11
|
|
|
@@ -159,6 +160,51 @@ function useRealtimeAvatarAudioSession(enabled) {
|
|
|
159
160
|
};
|
|
160
161
|
}, [enabled]);
|
|
161
162
|
}
|
|
163
|
+
var INPUT_LIVE_WAIT_MS = 5e3;
|
|
164
|
+
var INPUT_LIVE_POLL_MS = 100;
|
|
165
|
+
async function clearStaleInputMute(stillWanted) {
|
|
166
|
+
if (Platform.OS !== "ios") return;
|
|
167
|
+
try {
|
|
168
|
+
if (AudioDeviceModule.getMuteMode() !== AudioEngineMuteMode.VoiceProcessing) return;
|
|
169
|
+
for (const deadline = Date.now() + INPUT_LIVE_WAIT_MS; !AudioDeviceModule.isRecording(); ) {
|
|
170
|
+
if (Date.now() >= deadline || !stillWanted()) return;
|
|
171
|
+
await new Promise((resolve) => setTimeout(resolve, INPUT_LIVE_POLL_MS));
|
|
172
|
+
}
|
|
173
|
+
if (!stillWanted()) return;
|
|
174
|
+
try {
|
|
175
|
+
await AudioDeviceModule.setMicrophoneMuted(true);
|
|
176
|
+
} finally {
|
|
177
|
+
await AudioDeviceModule.setMicrophoneMuted(false);
|
|
178
|
+
}
|
|
179
|
+
} catch {
|
|
180
|
+
}
|
|
181
|
+
}
|
|
182
|
+
function useClearStaleInputMute(room) {
|
|
183
|
+
useEffect(() => {
|
|
184
|
+
if (!room) return;
|
|
185
|
+
let mounted = true;
|
|
186
|
+
const local = room.localParticipant;
|
|
187
|
+
const clearIfLive = (publication) => {
|
|
188
|
+
if (publication?.source !== Track.Source.Microphone || publication.isMuted) return;
|
|
189
|
+
void clearStaleInputMute(
|
|
190
|
+
() => mounted && !publication.isMuted && local.getTrackPublication(Track.Source.Microphone) === publication
|
|
191
|
+
);
|
|
192
|
+
};
|
|
193
|
+
const onUnmuted = (publication, participant) => {
|
|
194
|
+
if (participant === local) clearIfLive(publication);
|
|
195
|
+
};
|
|
196
|
+
room.on(RoomEvent.LocalTrackPublished, clearIfLive).on(RoomEvent.TrackUnmuted, onUnmuted);
|
|
197
|
+
clearIfLive(local.getTrackPublication(Track.Source.Microphone));
|
|
198
|
+
return () => {
|
|
199
|
+
mounted = false;
|
|
200
|
+
room.off(RoomEvent.LocalTrackPublished, clearIfLive).off(RoomEvent.TrackUnmuted, onUnmuted);
|
|
201
|
+
};
|
|
202
|
+
}, [room]);
|
|
203
|
+
}
|
|
204
|
+
function StaleInputMuteClearer() {
|
|
205
|
+
useClearStaleInputMute(useMaybeRoomContext());
|
|
206
|
+
return null;
|
|
207
|
+
}
|
|
162
208
|
function RealtimeAvatarLiveKitRoom(props) {
|
|
163
209
|
const {
|
|
164
210
|
grant,
|
|
@@ -186,6 +232,7 @@ function RealtimeAvatarLiveKitRoom(props) {
|
|
|
186
232
|
video: cameraPublishOptions(grant, video),
|
|
187
233
|
options: roomOptions
|
|
188
234
|
},
|
|
235
|
+
createElement(StaleInputMuteClearer),
|
|
189
236
|
children
|
|
190
237
|
);
|
|
191
238
|
}
|
|
@@ -3885,4 +3932,4 @@ function createProxyClient(options) {
|
|
|
3885
3932
|
};
|
|
3886
3933
|
}
|
|
3887
3934
|
|
|
3888
|
-
export { AdaptivePlayoutController, AvatarVideoSurface, DEFAULT_APPROACHING_END_LEAD_SECONDS, DEFAULT_AVATAR_PLAYOUT_DELAY_SECONDS, DEFAULT_CREDITS_LOW_LEAD_SECONDS, DEFAULT_FISH_TTS_MODEL, DEFAULT_GOVERNOR_CONFIG, DEFAULT_GRACE_CEILING_SECONDS, DEFAULT_GRACE_WINDOW_LEAD_SECONDS, DEFAULT_IDLE_SECONDS, DEFAULT_IDLE_WARN_LEAD_SECONDS, DEFAULT_READY_TIMEOUT_SECONDS, DEFAULT_TURN_TIMEOUT_SECONDS, MAX_RECONNECT_ATTEMPTS, MAX_SESSION_INSTRUCTIONS_CHARS, RECONNECT_BACKOFF_MS, RealtimeAvatarLiveKitRoom, SessionLifecycleRoomBridge, capacityErrorFromBusy, capacityStateFromGrant, createProxyClient, fishTtsModelSchema, isNativeLiveTrackSubscribed, knownBehaviorStates, mapTurnState, readInboundRtp, sessionBehaviorSchema, sessionClipSchema, splitCallTranscript, useAvatarAdaptivePlayoutDelay, useAvatarCamera, useAvatarPlayoutDelay, useAvatarQualityGovernor, useCallTranscript, useLiveKitAvatarGrant, useLiveTrackProducing, useMicLease, useRealtimeAvatarAudioSession, useRealtimeSession, useReleaseMicLeaseOnTrackEnded, useSessionLifecycle };
|
|
3935
|
+
export { AdaptivePlayoutController, AvatarVideoSurface, DEFAULT_APPROACHING_END_LEAD_SECONDS, DEFAULT_AVATAR_PLAYOUT_DELAY_SECONDS, DEFAULT_CREDITS_LOW_LEAD_SECONDS, DEFAULT_FISH_TTS_MODEL, DEFAULT_GOVERNOR_CONFIG, DEFAULT_GRACE_CEILING_SECONDS, DEFAULT_GRACE_WINDOW_LEAD_SECONDS, DEFAULT_IDLE_SECONDS, DEFAULT_IDLE_WARN_LEAD_SECONDS, DEFAULT_READY_TIMEOUT_SECONDS, DEFAULT_TURN_TIMEOUT_SECONDS, MAX_RECONNECT_ATTEMPTS, MAX_SESSION_INSTRUCTIONS_CHARS, RECONNECT_BACKOFF_MS, RealtimeAvatarLiveKitRoom, SessionLifecycleRoomBridge, capacityErrorFromBusy, capacityStateFromGrant, createProxyClient, fishTtsModelSchema, isNativeLiveTrackSubscribed, knownBehaviorStates, mapTurnState, readInboundRtp, sessionBehaviorSchema, sessionClipSchema, splitCallTranscript, useAvatarAdaptivePlayoutDelay, useAvatarCamera, useAvatarPlayoutDelay, useAvatarQualityGovernor, useCallTranscript, useClearStaleInputMute, useLiveKitAvatarGrant, useLiveTrackProducing, useMicLease, useRealtimeAvatarAudioSession, useRealtimeSession, useReleaseMicLeaseOnTrackEnded, useSessionLifecycle };
|
package/dist/server.d.ts
CHANGED
|
@@ -1,4 +1,4 @@
|
|
|
1
1
|
export { DEFAULT_FISH_TTS_MODEL, FishTtsModel, ListRendersQuery, ListRendersResponse, RealtimeAvatar, RealtimeAvatarError, RealtimeAvatarHttpError, RealtimeAvatarOptions, RealtimeAvatarWebhookError, Render, RenderAccessResponse, RenderFailureCode, RenderPlayback, RenderStatus, ShotTiming, StartCallOptions, WebhookVerificationErrorCode, clipLibraryDeclarationSchema, fishTtsModelSchema, listRendersResponseSchema, renderAccessResponseSchema, renderSchema, verifyTranscript, verifyVoiceWebhook } from './index.js';
|
|
2
|
-
export { h as Asset, g as AssetKind, A as Avatar, p as AvatarClip, c as AvatarSourceSwap, b as AvatarUpdate, q as CallConnection, C as CallMode, a as CallPolicy, r as CallQueued, s as ClipDeclaration, f as ClipLibrary, d as ClipLibraryDeclaration, t as ClipLibraryPlan, e as ClipLibraryUpdate, u as ClipSource, v as ContextMessage, n as CreateRenderInput, o as CreditBalance, E as EndCallOptions, w as EndCallReason, x as ExternalLlm, k as ListRecordingsQuery, l as ListRecordingsResponse, i as ListSessionsOptions, L as LoopRedirect, m as RecordingAccessResponse, R as RecordingArtifact, y as RecordingArtifactStatus, z as RecordingMode, B as RenderShot, S as StartCallResult, T as TranscriptPayload, j as UsageSession, U as UsageSessionPage, D as VideoPolicy, F as VoiceWebhook, G as VoiceWebhookErrorCode, V as VoiceWebhookRequest, H as isQueued, I as listRecordingsResponseSchema, J as recordingAccessResponseSchema, K as recordingArtifactSchema, M as recordingModeSchema } from './types-
|
|
2
|
+
export { h as Asset, g as AssetKind, A as Avatar, p as AvatarClip, c as AvatarSourceSwap, b as AvatarUpdate, q as CallConnection, C as CallMode, a as CallPolicy, r as CallQueued, s as ClipDeclaration, f as ClipLibrary, d as ClipLibraryDeclaration, t as ClipLibraryPlan, e as ClipLibraryUpdate, u as ClipSource, v as ContextMessage, n as CreateRenderInput, o as CreditBalance, E as EndCallOptions, w as EndCallReason, x as ExternalLlm, k as ListRecordingsQuery, l as ListRecordingsResponse, i as ListSessionsOptions, L as LoopRedirect, m as RecordingAccessResponse, R as RecordingArtifact, y as RecordingArtifactStatus, z as RecordingMode, B as RenderShot, S as StartCallResult, T as TranscriptPayload, j as UsageSession, U as UsageSessionPage, D as VideoPolicy, F as VoiceWebhook, G as VoiceWebhookErrorCode, V as VoiceWebhookRequest, H as isQueued, I as listRecordingsResponseSchema, J as recordingAccessResponseSchema, K as recordingArtifactSchema, M as recordingModeSchema } from './types-BSSqsu_Z.js';
|
|
3
3
|
import './connection-history.js';
|
|
4
4
|
import 'zod';
|
package/dist/server.js
CHANGED
|
@@ -1,2 +1,2 @@
|
|
|
1
|
-
export { DEFAULT_FISH_TTS_MODEL, RealtimeAvatar, RealtimeAvatarError, RealtimeAvatarHttpError, RealtimeAvatarWebhookError, clipLibraryDeclarationSchema, fishTtsModelSchema, isQueued, listRecordingsResponseSchema, listRendersResponseSchema, recordingAccessResponseSchema, recordingArtifactSchema, recordingModeSchema, renderAccessResponseSchema, renderSchema, verifyTranscript, verifyVoiceWebhook } from './chunk-
|
|
1
|
+
export { DEFAULT_FISH_TTS_MODEL, RealtimeAvatar, RealtimeAvatarError, RealtimeAvatarHttpError, RealtimeAvatarWebhookError, clipLibraryDeclarationSchema, fishTtsModelSchema, isQueued, listRecordingsResponseSchema, listRendersResponseSchema, recordingAccessResponseSchema, recordingArtifactSchema, recordingModeSchema, renderAccessResponseSchema, renderSchema, verifyTranscript, verifyVoiceWebhook } from './chunk-NKS5ZIWN.js';
|
|
2
2
|
import './chunk-MCIXU6XP.js';
|
package/dist/tanstack-start.d.ts
CHANGED
package/dist/tanstack-start.js
CHANGED
|
@@ -90,6 +90,11 @@ interface components {
|
|
|
90
90
|
speed?: number | null;
|
|
91
91
|
emotion?: string | null;
|
|
92
92
|
language?: string | null;
|
|
93
|
+
whisper?: {
|
|
94
|
+
/** @enum {string} */
|
|
95
|
+
model: "s2-pro" | "s2.1-pro" | "drama-3";
|
|
96
|
+
gain_db?: number;
|
|
97
|
+
} | null;
|
|
93
98
|
}) | null;
|
|
94
99
|
voice_id?: string | null;
|
|
95
100
|
/** @description The session's motion-state map. Normally omitted: the avatar's PLATFORM-GENERATED library (idle / listen / gesture clips rendered from the portrait at creation) is injected automatically. An explicit [] opts the call out of clips. Supplying custom clip URLs is a legacy channel for existing integrations — custom video is not part of the public contract. */
|
|
@@ -207,7 +212,7 @@ interface components {
|
|
|
207
212
|
/** @description DEPRECATED, no-op. Accepted and ignored. The faster hesitation ceiling this requested is now the default for every session, so sending it changes nothing. Safe to stop sending; the field is retained only so existing callers do not break. */
|
|
208
213
|
fast_endpointing?: boolean;
|
|
209
214
|
require_ready_clip_library?: boolean;
|
|
210
|
-
/** @description Preview, enabled per workspace. A prepared scene from POST /v1/scenes, played forward over its full length in place of the avatar's source video; the avatar keeps the voice and persona. Cannot be combined with mode voice, render_backend generative, source_video_url, video_cache_id or a non-empty clip_library. */
|
|
215
|
+
/** @description Preview, enabled per workspace. A prepared scene from POST /v1/scenes, played forward over its full length in place of the avatar's source video; the avatar keeps the voice and persona. Cannot be combined with mode voice, render_backend generative, source_video_url, video_cache_id or a non-empty clip_library (unless scene.clips is library, whose clips are scenes this workspace prepared). */
|
|
211
216
|
scene?: {
|
|
212
217
|
/** @description The prepared scene's artifact id, as returned by POST /v1/scenes when status is ready. */
|
|
213
218
|
video_cache_id: string;
|
|
@@ -219,6 +224,26 @@ interface components {
|
|
|
219
224
|
attach_continuation?: boolean;
|
|
220
225
|
/** @description Preview, opt-in. Lets the minting server switch this scene session to another ready scene it owns mid-call, without a new session. Requires extend_secret to sign replace_scene commands. Absent or false plays this scene for the whole call. */
|
|
221
226
|
replaceable?: boolean;
|
|
227
|
+
/**
|
|
228
|
+
* @description Preview, opt-in. How the speaking render is fitted onto the filmed scene: frame keeps the lips on each filmed frame's face (its own position and mouth). Absent renders as before.
|
|
229
|
+
* @enum {string}
|
|
230
|
+
*/
|
|
231
|
+
speech_fit?: "frame";
|
|
232
|
+
/**
|
|
233
|
+
* @description Preview, opt-in. Which part of the face the speaking render replaces on the filmed scene: full re-renders the whole face, as a video call does. Absent re-renders the mouth and jaw only, as before.
|
|
234
|
+
* @enum {string}
|
|
235
|
+
*/
|
|
236
|
+
speech_mask?: "full";
|
|
237
|
+
/**
|
|
238
|
+
* @description Preview, opt-in. library lets this scene session carry clip_library and clip_behavior: gesture clips that are scenes this workspace prepared, each starting on this scene's first frame and ending on its last. The character picks them as reactions and gestures; they play at the scene's loop boundary. A clip that cannot be admitted is left out, never refused. Absent plays this scene alone.
|
|
239
|
+
* @enum {string}
|
|
240
|
+
*/
|
|
241
|
+
clips?: "library";
|
|
242
|
+
/**
|
|
243
|
+
* @description Preview, opt-in, needs clips. hurry plays the scene at twice its speed to its last frame while a gesture waits, so the gesture starts sooner; hurry_silent does so only while she is silent. Absent waits for the scene's natural end.
|
|
244
|
+
* @enum {string}
|
|
245
|
+
*/
|
|
246
|
+
clip_lead?: "hurry" | "hurry_silent";
|
|
222
247
|
};
|
|
223
248
|
};
|
|
224
249
|
LiveKitSessionGrant: {
|
|
@@ -1267,6 +1292,11 @@ interface components {
|
|
|
1267
1292
|
speed?: number | null;
|
|
1268
1293
|
emotion?: string | null;
|
|
1269
1294
|
language?: string | null;
|
|
1295
|
+
whisper?: {
|
|
1296
|
+
/** @enum {string} */
|
|
1297
|
+
model: "s2-pro" | "s2.1-pro" | "drama-3";
|
|
1298
|
+
gain_db?: number;
|
|
1299
|
+
} | null;
|
|
1270
1300
|
};
|
|
1271
1301
|
/** @description Mouth-motion amplitude, as on a live call (lip_gain). Omit it for the platform default. */
|
|
1272
1302
|
lipGain?: number;
|
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
import { C as CallMode, a as CallPolicy } from './types-
|
|
1
|
+
import { C as CallMode, a as CallPolicy } from './types-BSSqsu_Z.js';
|
|
2
2
|
|
|
3
3
|
/** The three things a caller can ask the proxy to do. Gate on these, not on URLs. */
|
|
4
4
|
type ProxyOperation = "connect" | "end" | "avatars" | "credits";
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "realtime-avatar",
|
|
3
|
-
"version": "0.
|
|
3
|
+
"version": "0.32.0",
|
|
4
4
|
"description": "Realtime Avatar SDK. `realtime-avatar` and the route adapters hold your API key and are server-only; `realtime-avatar/react`, `/browser` and `/tools` never can — importing a server entry into a browser build throws at bundle time.",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"author": "The Influence Company",
|