realtime-avatar 0.27.5 → 0.28.0
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 +48 -2
- package/dist/{chunk-IJHM44YT.js → chunk-K3VC6BF7.js} +2 -1
- package/dist/{chunk-6E3SZH35.js → chunk-V7A4NRKC.js} +177 -62
- 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 +36 -6
- package/dist/index.js +1 -1
- package/dist/nextjs.d.ts +2 -2
- package/dist/nextjs.js +2 -2
- package/dist/{proxy-client-B7BAxMyL.d.ts → proxy-client-WjGIwL-t.d.ts} +44 -1
- package/dist/react-native.d.ts +2 -2
- package/dist/react-native.js +50 -4
- package/dist/react.d.ts +12 -3
- package/dist/react.js +54 -6
- package/dist/server.d.ts +2 -2
- package/dist/server.js +1 -1
- package/dist/tanstack-start.d.ts +2 -2
- package/dist/tanstack-start.js +2 -2
- package/dist/{types-B1IR4EaJ.d.ts → types-Bli1UeP0.d.ts} +1 -1
- package/dist/{types-C8X50QT9.d.ts → types-yiemDkm0.d.ts} +274 -5
- package/package.json +1 -1
|
@@ -184,6 +184,27 @@ interface components {
|
|
|
184
184
|
client_metadata?: {
|
|
185
185
|
[key: string]: string;
|
|
186
186
|
};
|
|
187
|
+
/** @description Speak this call with your own voice: the worker POSTs each sentence to url and plays the audio you return. Replaces voice and voice_id. */
|
|
188
|
+
voice_webhook?: {
|
|
189
|
+
/**
|
|
190
|
+
* Format: uri
|
|
191
|
+
* @description https:// endpoint on your server that returns speech audio for one sentence. The worker also refuses a host that resolves to a non-public address and never follows redirects.
|
|
192
|
+
*/
|
|
193
|
+
url: string;
|
|
194
|
+
/** @description HMAC-SHA256 key for x-rta-signature, the same scheme as transcript_webhook. Never placed in room or participant metadata. */
|
|
195
|
+
secret: string;
|
|
196
|
+
/** @description Your own voice identifier, echoed in every request as voice_id. */
|
|
197
|
+
voice_id?: string;
|
|
198
|
+
/** @description BCP-47 language hint, echoed in every request as language. */
|
|
199
|
+
language?: string;
|
|
200
|
+
/**
|
|
201
|
+
* @description Sample rate of raw audio/pcm replies, in Hz. Default 24000.
|
|
202
|
+
* @enum {number}
|
|
203
|
+
*/
|
|
204
|
+
sample_rate?: 8000 | 16000 | 22050 | 24000 | 44100 | 48000;
|
|
205
|
+
/** @description In milliseconds, default 5000. Bounds the first audio byte of each attempt and every later gap between chunks; past it the reply fails with voice_webhook_timeout. */
|
|
206
|
+
timeout_ms?: number;
|
|
207
|
+
};
|
|
187
208
|
/** @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. */
|
|
188
209
|
fast_endpointing?: boolean;
|
|
189
210
|
require_ready_clip_library?: boolean;
|
|
@@ -655,6 +676,48 @@ interface components {
|
|
|
655
676
|
} & {
|
|
656
677
|
[key: string]: unknown;
|
|
657
678
|
};
|
|
679
|
+
/** @description One sentence to synthesize, signed with voice_webhook.secret. Reply 200 with streamed audio. */
|
|
680
|
+
VoiceWebhookRequest: {
|
|
681
|
+
/** @constant */
|
|
682
|
+
type: "voice.synthesize";
|
|
683
|
+
/** @constant */
|
|
684
|
+
version: 1;
|
|
685
|
+
/** @description Unique per POST; also sent as the x-rta-request-id header. */
|
|
686
|
+
request_id: string;
|
|
687
|
+
/** @description The same session id the transcript webhook payload carries. */
|
|
688
|
+
session_id: string;
|
|
689
|
+
/** @description The plain words to speak: no markup and no stage directions. */
|
|
690
|
+
text: string;
|
|
691
|
+
/** @description voice_webhook.voice_id from the mint, or null. */
|
|
692
|
+
voice_id: string | null;
|
|
693
|
+
/** @description voice_webhook.language from the mint, or null. */
|
|
694
|
+
language: string | null;
|
|
695
|
+
/** @description The raw audio/pcm format expected when the reply does not name another. */
|
|
696
|
+
output: {
|
|
697
|
+
/** @constant */
|
|
698
|
+
encoding: "pcm_s16le";
|
|
699
|
+
/** @enum {number} */
|
|
700
|
+
sample_rate: 8000 | 16000 | 22050 | 24000 | 44100 | 48000;
|
|
701
|
+
/** @constant */
|
|
702
|
+
channels: 1;
|
|
703
|
+
};
|
|
704
|
+
/** @description Mint-time client_metadata echoed verbatim; an empty object when none was provided. */
|
|
705
|
+
client_metadata: {
|
|
706
|
+
[key: string]: string;
|
|
707
|
+
};
|
|
708
|
+
};
|
|
709
|
+
/** @enum {string} */
|
|
710
|
+
VoiceWebhookErrorCode: "voice_webhook_url_not_allowed" | "voice_webhook_unauthorized" | "voice_webhook_unreachable" | "voice_webhook_timeout" | "voice_webhook_http_error" | "voice_webhook_unsupported_audio" | "voice_webhook_empty_audio";
|
|
711
|
+
VoiceErrorFrame: {
|
|
712
|
+
/** @constant */
|
|
713
|
+
kind: "voice_error";
|
|
714
|
+
/** @enum {string} */
|
|
715
|
+
code: "voice_webhook_url_not_allowed" | "voice_webhook_unauthorized" | "voice_webhook_unreachable" | "voice_webhook_timeout" | "voice_webhook_http_error" | "voice_webhook_unsupported_audio" | "voice_webhook_empty_audio";
|
|
716
|
+
message: string;
|
|
717
|
+
status?: number;
|
|
718
|
+
fatal: boolean;
|
|
719
|
+
request_id?: string;
|
|
720
|
+
};
|
|
658
721
|
Avatar: {
|
|
659
722
|
id: string;
|
|
660
723
|
tenantId: string;
|
|
@@ -774,7 +837,7 @@ interface components {
|
|
|
774
837
|
* "avatars:read"
|
|
775
838
|
* ]
|
|
776
839
|
*/
|
|
777
|
-
scopes: ("*" | "api_keys:write" | "credits:read" | "avatars:read" | "avatars:write" | "realtime:write" | "usage:read" | "recordings:read" | "usage:write")[];
|
|
840
|
+
scopes: ("*" | "api_keys:write" | "credits:read" | "avatars:read" | "avatars:write" | "realtime:write" | "usage:read" | "recordings:read" | "usage:write" | "speech:write")[];
|
|
778
841
|
spendLimitCreditMicros?: number | null;
|
|
779
842
|
expiresAt?: string | null;
|
|
780
843
|
};
|
|
@@ -786,7 +849,7 @@ interface components {
|
|
|
786
849
|
redactedKey: string;
|
|
787
850
|
/** @default false */
|
|
788
851
|
canReveal: boolean;
|
|
789
|
-
scopes: ("*" | "api_keys:write" | "credits:read" | "avatars:read" | "avatars:write" | "realtime:write" | "usage:read" | "recordings:read" | "usage:write")[];
|
|
852
|
+
scopes: ("*" | "api_keys:write" | "credits:read" | "avatars:read" | "avatars:write" | "realtime:write" | "usage:read" | "recordings:read" | "usage:write" | "speech:write")[];
|
|
790
853
|
/** @enum {string} */
|
|
791
854
|
status: "active" | "revoked" | "expired";
|
|
792
855
|
spendLimitCreditMicros: number | null;
|
|
@@ -898,6 +961,20 @@ interface components {
|
|
|
898
961
|
total_ms?: number;
|
|
899
962
|
};
|
|
900
963
|
};
|
|
964
|
+
CreateSpeechRequest: {
|
|
965
|
+
/** @description What to say, at most 2,000 characters after trimming. Billed per UTF-8 byte: plain English is one byte per character, accented letters and other scripts use 2 to 4. Text in [square brackets] may be performed as a delivery cue rather than spoken. */
|
|
966
|
+
text: string;
|
|
967
|
+
/**
|
|
968
|
+
* @description mp3 (44.1 kHz, 128 kbps, mono) or wav.
|
|
969
|
+
* @default mp3
|
|
970
|
+
* @enum {string}
|
|
971
|
+
*/
|
|
972
|
+
format: "mp3" | "wav";
|
|
973
|
+
/** @description Speaking rate, 0.5 to 2. Omit for the voice's natural pace. */
|
|
974
|
+
speed?: number;
|
|
975
|
+
/** @description BCP-47 tag such as en-AU. Shapes how numbers, dates and units are read; it does not change the voice. */
|
|
976
|
+
language?: string;
|
|
977
|
+
};
|
|
901
978
|
CreditBalance: {
|
|
902
979
|
tenantId: string;
|
|
903
980
|
balanceCreditMicros: number;
|
|
@@ -1400,9 +1477,9 @@ interface components {
|
|
|
1400
1477
|
* @description Present on `403 insufficient_scope` — the one scope this operation needed and the key did not carry. Repeated in the `WWW-Authenticate` challenge as `scope=`.
|
|
1401
1478
|
* @enum {string}
|
|
1402
1479
|
*/
|
|
1403
|
-
requiredScope?: "*" | "api_keys:write" | "credits:read" | "avatars:read" | "avatars:write" | "realtime:write" | "usage:read" | "recordings:read" | "usage:write";
|
|
1480
|
+
requiredScope?: "*" | "api_keys:write" | "credits:read" | "avatars:read" | "avatars:write" | "realtime:write" | "usage:read" | "recordings:read" | "usage:write" | "speech:write";
|
|
1404
1481
|
/** @description Present alongside `requiredScope` — the scopes the presented key does carry. */
|
|
1405
|
-
grantedScopes?: ("*" | "api_keys:write" | "credits:read" | "avatars:read" | "avatars:write" | "realtime:write" | "usage:read" | "recordings:read" | "usage:write")[];
|
|
1482
|
+
grantedScopes?: ("*" | "api_keys:write" | "credits:read" | "avatars:read" | "avatars:write" | "realtime:write" | "usage:read" | "recordings:read" | "usage:write" | "speech:write")[];
|
|
1406
1483
|
/** @description Present on 402 — where to send the user to top up. */
|
|
1407
1484
|
billingUrl?: string;
|
|
1408
1485
|
/** @description On concurrency_limit_reached: the workspace's concurrent session limit. */
|
|
@@ -1709,6 +1786,102 @@ interface operations {
|
|
|
1709
1786
|
};
|
|
1710
1787
|
};
|
|
1711
1788
|
};
|
|
1789
|
+
createSpeech: {
|
|
1790
|
+
parameters: {
|
|
1791
|
+
query?: never;
|
|
1792
|
+
header?: never;
|
|
1793
|
+
path: {
|
|
1794
|
+
voiceId: string;
|
|
1795
|
+
};
|
|
1796
|
+
cookie?: never;
|
|
1797
|
+
};
|
|
1798
|
+
requestBody: {
|
|
1799
|
+
content: {
|
|
1800
|
+
"application/json": components["schemas"]["CreateSpeechRequest"];
|
|
1801
|
+
};
|
|
1802
|
+
};
|
|
1803
|
+
responses: {
|
|
1804
|
+
/** @description The audio file, in the requested format. */
|
|
1805
|
+
200: {
|
|
1806
|
+
headers: {
|
|
1807
|
+
/** @description Credits billed for this request. */
|
|
1808
|
+
"X-RTA-Billed-Credits"?: number;
|
|
1809
|
+
/** @description UTF-8 bytes of the trimmed text that were metered. */
|
|
1810
|
+
"X-RTA-Billed-Bytes"?: number;
|
|
1811
|
+
[name: string]: unknown;
|
|
1812
|
+
};
|
|
1813
|
+
content: {
|
|
1814
|
+
"audio/mpeg": string;
|
|
1815
|
+
"audio/wav": string;
|
|
1816
|
+
};
|
|
1817
|
+
};
|
|
1818
|
+
/** @description Missing, malformed, revoked, or expired key. */
|
|
1819
|
+
401: {
|
|
1820
|
+
headers: {
|
|
1821
|
+
[name: string]: unknown;
|
|
1822
|
+
};
|
|
1823
|
+
content: {
|
|
1824
|
+
"application/json": components["schemas"]["Error"];
|
|
1825
|
+
};
|
|
1826
|
+
};
|
|
1827
|
+
/** @description The key lacks the scope for this operation, the tenant is not active, or the workspace is not a paying one (code "speech_requires_purchase"). Speech is open to workspaces on a paid plan or that have bought credits; buying a credit pack unlocks it. */
|
|
1828
|
+
403: {
|
|
1829
|
+
headers: {
|
|
1830
|
+
[name: string]: unknown;
|
|
1831
|
+
};
|
|
1832
|
+
content: {
|
|
1833
|
+
"application/json": components["schemas"]["Error"];
|
|
1834
|
+
};
|
|
1835
|
+
};
|
|
1836
|
+
/** @description No voice with this id is available to this workspace (code "voice_not_found"). */
|
|
1837
|
+
404: {
|
|
1838
|
+
headers: {
|
|
1839
|
+
[name: string]: unknown;
|
|
1840
|
+
};
|
|
1841
|
+
content: {
|
|
1842
|
+
"application/json": components["schemas"]["Error"];
|
|
1843
|
+
};
|
|
1844
|
+
};
|
|
1845
|
+
/** @description The voice is still being prepared (code "voice_not_ready"). Retry shortly. */
|
|
1846
|
+
409: {
|
|
1847
|
+
headers: {
|
|
1848
|
+
[name: string]: unknown;
|
|
1849
|
+
};
|
|
1850
|
+
content: {
|
|
1851
|
+
"application/json": components["schemas"]["Error"];
|
|
1852
|
+
};
|
|
1853
|
+
};
|
|
1854
|
+
/** @description A request field or the voice id prefix is invalid; the voice cannot be used to generate speech (code "voice_unavailable"; choose another voice); the voice provider refused this request (code "speech_rejected"; check the text, format, speed and language); or the audio for this text is longer than one response can carry (code "speech_too_long"; split the text, or ask for mp3). None of these is retryable unchanged. */
|
|
1855
|
+
422: {
|
|
1856
|
+
headers: {
|
|
1857
|
+
[name: string]: unknown;
|
|
1858
|
+
};
|
|
1859
|
+
content: {
|
|
1860
|
+
"application/json": components["schemas"]["Error"];
|
|
1861
|
+
};
|
|
1862
|
+
};
|
|
1863
|
+
/** @description Per-workspace rate limit (code "rate_limited"), or speech generation is busy (code "speech_busy"): this workspace already has a request in progress, or the platform-wide cap is reached. Both are retryable; honour Retry-After. */
|
|
1864
|
+
429: {
|
|
1865
|
+
headers: {
|
|
1866
|
+
/** @description Seconds to wait before retrying. */
|
|
1867
|
+
"Retry-After"?: number;
|
|
1868
|
+
[name: string]: unknown;
|
|
1869
|
+
};
|
|
1870
|
+
content: {
|
|
1871
|
+
"application/json": components["schemas"]["Error"];
|
|
1872
|
+
};
|
|
1873
|
+
};
|
|
1874
|
+
/** @description Error. Route-dependent: 400 (a field this route cannot honour — see `portraitUrl` on `UpdateAvatarRequest`), 402 (insufficient credits or spend limit), 404 (no such avatar, voice, or key for this tenant — the message often reads "does not belong to this tenant", which is a missing id and not a permission failure), 408 (the request was cancelled), 409 (the resource is not in a state that accepts this write — a stale `expectedRevision`, a render already in flight, or a mint that asked for a clip library still building), 411 (`POST /v1/assets/remote` only — the origin serving `remoteUrl` sent no `content-length`, so the platform will not stream it), 422 (strict schema rejection, or a motion description the safety screen refused), 429 (rate limited), 502 (upstream render failed), 503 (a dependency this route needs is unavailable — retryable, and nothing was written), 504 (the upstream request budget expired), 500 (unhandled). Switch on `code` where present, else `status`. */
|
|
1875
|
+
default: {
|
|
1876
|
+
headers: {
|
|
1877
|
+
[name: string]: unknown;
|
|
1878
|
+
};
|
|
1879
|
+
content: {
|
|
1880
|
+
"application/json": components["schemas"]["Error"];
|
|
1881
|
+
};
|
|
1882
|
+
};
|
|
1883
|
+
};
|
|
1884
|
+
};
|
|
1712
1885
|
listAvatars: {
|
|
1713
1886
|
parameters: {
|
|
1714
1887
|
query?: never;
|
|
@@ -2786,6 +2959,68 @@ interface operations {
|
|
|
2786
2959
|
};
|
|
2787
2960
|
};
|
|
2788
2961
|
};
|
|
2962
|
+
synthesizeVoice: {
|
|
2963
|
+
parameters: {
|
|
2964
|
+
query?: never;
|
|
2965
|
+
header: {
|
|
2966
|
+
/** @description v1= followed by the hex HMAC-SHA256 of <timestamp>.<raw body>, keyed by voice_webhook.secret. Compare in constant time. */
|
|
2967
|
+
"x-rta-signature": string;
|
|
2968
|
+
/** @description Unix seconds, used verbatim in the signed message. Verify freshness to bound replay. */
|
|
2969
|
+
"x-rta-timestamp": string;
|
|
2970
|
+
/** @description Equal to request_id in the body. Unique per POST. */
|
|
2971
|
+
"x-rta-request-id": string;
|
|
2972
|
+
};
|
|
2973
|
+
path?: never;
|
|
2974
|
+
cookie?: never;
|
|
2975
|
+
};
|
|
2976
|
+
requestBody: {
|
|
2977
|
+
content: {
|
|
2978
|
+
"application/json": components["schemas"]["VoiceWebhookRequest"];
|
|
2979
|
+
};
|
|
2980
|
+
};
|
|
2981
|
+
responses: {
|
|
2982
|
+
/** @description Speech audio for request.text, streamed. */
|
|
2983
|
+
200: {
|
|
2984
|
+
headers: {
|
|
2985
|
+
[name: string]: unknown;
|
|
2986
|
+
};
|
|
2987
|
+
content: {
|
|
2988
|
+
"audio/pcm": string;
|
|
2989
|
+
"audio/wav": string;
|
|
2990
|
+
"audio/mpeg": string;
|
|
2991
|
+
"audio/ogg": string;
|
|
2992
|
+
};
|
|
2993
|
+
};
|
|
2994
|
+
/** @description No audio: voice_webhook_empty_audio, as is any 2xx with no body. The rest of that reply is dropped. */
|
|
2995
|
+
204: {
|
|
2996
|
+
headers: {
|
|
2997
|
+
[name: string]: unknown;
|
|
2998
|
+
};
|
|
2999
|
+
content?: never;
|
|
3000
|
+
};
|
|
3001
|
+
/** @description Signature rejected. Ends the call with voice_webhook_unauthorized. */
|
|
3002
|
+
401: {
|
|
3003
|
+
headers: {
|
|
3004
|
+
[name: string]: unknown;
|
|
3005
|
+
};
|
|
3006
|
+
content?: never;
|
|
3007
|
+
};
|
|
3008
|
+
/** @description Refused. Ends the call with voice_webhook_unauthorized. */
|
|
3009
|
+
403: {
|
|
3010
|
+
headers: {
|
|
3011
|
+
[name: string]: unknown;
|
|
3012
|
+
};
|
|
3013
|
+
content?: never;
|
|
3014
|
+
};
|
|
3015
|
+
/** @description Any other status, a 3xx included, is voice_webhook_http_error. Statuses 408, 429 and 5xx are retried once, as a new POST with a new request_id; any other status, or a failed retry, drops the rest of that reply. */
|
|
3016
|
+
default: {
|
|
3017
|
+
headers: {
|
|
3018
|
+
[name: string]: unknown;
|
|
3019
|
+
};
|
|
3020
|
+
content?: never;
|
|
3021
|
+
};
|
|
3022
|
+
};
|
|
3023
|
+
};
|
|
2789
3024
|
}
|
|
2790
3025
|
|
|
2791
3026
|
declare const listUsageSessionsResponseSchema: z.ZodObject<{
|
|
@@ -3126,6 +3361,17 @@ declare const recordingAccessResponseSchema: z.ZodObject<{
|
|
|
3126
3361
|
}, z.core.$strict>;
|
|
3127
3362
|
type RecordingAccessResponse = z.infer<typeof recordingAccessResponseSchema>;
|
|
3128
3363
|
|
|
3364
|
+
type Wire$1 = components["schemas"];
|
|
3365
|
+
declare const voiceWebhookOptionsSchema: z.ZodObject<{
|
|
3366
|
+
url: z.ZodURL;
|
|
3367
|
+
secret: z.ZodString;
|
|
3368
|
+
voiceId: z.ZodOptional<z.ZodString>;
|
|
3369
|
+
language: z.ZodOptional<z.ZodString>;
|
|
3370
|
+
sampleRate: z.ZodOptional<z.ZodLiteral<8000 | 16000 | 22050 | 24000 | 44100 | 48000>>;
|
|
3371
|
+
timeoutMs: z.ZodOptional<z.ZodInt>;
|
|
3372
|
+
}, z.core.$strict>;
|
|
3373
|
+
declare function voiceWebhookToWire(input: z.input<typeof voiceWebhookOptionsSchema>): NonNullable<Wire$1["LiveKitSessionRequest"]["voice_webhook"]>;
|
|
3374
|
+
|
|
3129
3375
|
/**
|
|
3130
3376
|
* The public shapes. Everything an integrator touches is here, in one file, on purpose:
|
|
3131
3377
|
* an agent reading this repo should be able to learn the whole surface without following
|
|
@@ -3285,6 +3531,24 @@ interface ExternalLlm {
|
|
|
3285
3531
|
/** Sent as `model`. Omit it to let your endpoint choose. */
|
|
3286
3532
|
model?: NonNullable<WireLlm["model"]>;
|
|
3287
3533
|
}
|
|
3534
|
+
/**
|
|
3535
|
+
* Speak this call with your own voice. Each sentence she is about to say is POSTed to `url`,
|
|
3536
|
+
* signed with `secret` (verify it with `verifyVoiceWebhook`), and the audio you stream back is what
|
|
3537
|
+
* she says, lip-synced in both modes.
|
|
3538
|
+
*
|
|
3539
|
+
* `url` must be https and resolve to a public address; redirects are never followed. `secret` is
|
|
3540
|
+
* 16–200 characters and never reaches room or participant metadata. `voiceId` and `language` are
|
|
3541
|
+
* echoed in every request. `sampleRate` (default 24000) is the rate of raw `audio/pcm` replies.
|
|
3542
|
+
* `timeoutMs` (1000–15000, default 5000) bounds the first audio byte of each attempt and every
|
|
3543
|
+
* later gap between chunks.
|
|
3544
|
+
*
|
|
3545
|
+
* It replaces the built-in voice: a mint carrying `voice` as well is refused with 422.
|
|
3546
|
+
*/
|
|
3547
|
+
type VoiceWebhook = Parameters<typeof voiceWebhookToWire>[0];
|
|
3548
|
+
/** The signed `voice.synthesize` request your `VoiceWebhook.url` receives, as `verifyVoiceWebhook` returns it. */
|
|
3549
|
+
type VoiceWebhookRequest = Wire["VoiceWebhookRequest"];
|
|
3550
|
+
/** Why your voice webhook failed: the `code` of an in-call voice error. */
|
|
3551
|
+
type VoiceWebhookErrorCode = Wire["VoiceWebhookErrorCode"];
|
|
3288
3552
|
/** What YOUR SERVER decides about a call. Never accept any of this from a browser. */
|
|
3289
3553
|
interface CallPolicy {
|
|
3290
3554
|
/** Optional server-owned content safety policy. Never accept a selection or disable from the browser. */
|
|
@@ -3312,6 +3576,11 @@ interface CallPolicy {
|
|
|
3312
3576
|
video?: VideoPolicy;
|
|
3313
3577
|
/** Voice override for this call; omit to use the avatar's default. */
|
|
3314
3578
|
voice?: unknown;
|
|
3579
|
+
/**
|
|
3580
|
+
* Speak with your own voice: your endpoint answers each sentence with audio. Carries a secret,
|
|
3581
|
+
* so like `llm` the browser entries of this package cannot send it. See {@link VoiceWebhook}.
|
|
3582
|
+
*/
|
|
3583
|
+
voiceWebhook?: VoiceWebhook;
|
|
3315
3584
|
/**
|
|
3316
3585
|
* Let the browser register tools the model may call.
|
|
3317
3586
|
*
|
|
@@ -3565,4 +3834,4 @@ interface TranscriptPayload {
|
|
|
3565
3834
|
client_metadata: Record<string, string>;
|
|
3566
3835
|
}
|
|
3567
3836
|
|
|
3568
|
-
export { type Avatar as A,
|
|
3837
|
+
export { type Avatar as A, type VoiceWebhook as B, type CallMode as C, type VoiceWebhookErrorCode as D, type EndCallOptions as E, isQueued as F, listRecordingsResponseSchema as G, recordingAccessResponseSchema as H, recordingArtifactSchema as I, recordingModeSchema as J, type LoopRedirect as L, type RecordingArtifact as R, type StartCallResult as S, type TranscriptPayload as T, type UsageSessionPage as U, type VoiceWebhookRequest as V, type CallPolicy as a, type AvatarUpdate as b, type AvatarSourceSwap as c, type ClipLibraryDeclaration as d, type ClipLibraryUpdate as e, type ClipLibrary as f, type AssetKind as g, type Asset as h, type ListSessionsOptions as i, type UsageSession as j, type ListRecordingsQuery as k, type ListRecordingsResponse as l, type RecordingAccessResponse as m, type CreditBalance as n, type AvatarClip as o, type CallConnection as p, type CallQueued as q, type ClipDeclaration as r, type ClipLibraryPlan as s, type ClipSource as t, type ContextMessage as u, type EndCallReason as v, type ExternalLlm as w, type RecordingArtifactStatus as x, type RecordingMode as y, type VideoPolicy as z };
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "realtime-avatar",
|
|
3
|
-
"version": "0.
|
|
3
|
+
"version": "0.28.0",
|
|
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",
|