@wildix/wilma-voice-client 0.0.38 → 0.0.39

package/dist-cjs/models/enums.js

@@ -1,6 +1,6 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.VoiceCommandType = exports.VoiceInputMode = exports.VoiceAudioEncoding = exports.VoiceCallToolPhase = exports.VoiceCallEndReason = exports.VoiceControlAction = exports.VoiceEventType = exports.VoicePronunciationEncoding = void 0;
+exports.VoiceCommandType = exports.VoiceInputMode = exports.VoiceAudioEncoding = exports.VoiceCallToolPhase = exports.VoiceCallEndReason = exports.VoiceCallBackendTransport = exports.VoiceControlAction = exports.VoiceEventType = exports.VoicePronunciationEncoding = void 0;
 exports.VoicePronunciationEncoding = {
     ALIAS: "alias",
     IPA: "ipa",
@@ -23,6 +23,10 @@ exports.VoiceControlAction = {
     END_OF_TURN: "end_of_turn",
     KEEPALIVE: "keepalive",
 };
+exports.VoiceCallBackendTransport = {
+    SQS: "sqs",
+    WEBHOOK: "webhook",
+};
 exports.VoiceCallEndReason = {
     COMPLETE: "complete",
     ERROR: "error",

package/dist-cjs/schemas/schemas_0.js

@@ -1,7 +1,7 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.StreamCallEventsOutput$ = exports.StreamCallEventsInput$ = exports.StopWebsocketCommand$ = exports.StopCommandData$ = exports.SessionStartedWebsocketEvent$ = exports.SessionStartedEvent$ = exports.SessionEndedWebsocketEvent$ = exports.SessionEndedEvent$ = exports.SendTransferOutput$ = exports.SendTransferInput$ = exports.SendStopOutput$ = exports.SendStopInput$ = exports.SendSayOutput$ = exports.SendSayInput$ = exports.SendPlayOutput$ = exports.SendPlayInput$ = exports.SendHangupOutput$ = exports.SendHangupInput$ = exports.SayWebsocketCommand$ = exports.SayCommandData$ = exports.SayCommandAudioOptions$ = exports.ReplyWebsocketEvent$ = exports.ReplyEvent$ = exports.PlayWebsocketCommand$ = exports.PlayCommandData$ = exports.PlaybackWebsocketEvent$ = exports.PlaybackEvent$ = exports.ListVoicesOutput$ = exports.ListVoicesInput$ = exports.InterruptionWebsocketEvent$ = exports.InterruptionEvent$ = exports.HangupWebsocketCommand$ = exports.HangupCommandData$ = exports.ErrorWebsocketEvent$ = exports.DescribeVoiceProtocolOutput$ = exports.DescribeVoiceProtocolInput$ = exports.ControlWebsocketEvent$ = exports.ControlEvent$ = exports.CompleteWebsocketEvent$ = exports.CompleteEvent$ = exports.AudioWebsocketEvent$ = exports.AudioEvent$ = exports.AgentTranscriptionWebsocketEvent$ = exports.AgentTranscriptionEvent$ = exports.VoicePronunciation$ = exports.errorTypeRegistries = exports.ValidationException$ = exports.NotFoundException$ = exports.ForbiddenException$ = exports.WilmaVoiceServiceException$ = void 0;
-exports.TextToSpeech$ = exports.StreamCallEvents$ = exports.SendTransfer$ = exports.SendStop$ = exports.SendSay$ = exports.SendPlay$ = exports.SendHangup$ = exports.ListVoices$ = exports.DescribeVoiceProtocol$ = exports.VoiceCallEvent$ = exports.VoiceSpeechPayload$ = exports.VoiceErrorEvent$ = exports.VoiceConnectParams$ = exports.VoiceCallTranscriptionEvent$ = exports.VoiceCallToolEvent$ = exports.VoiceCallStartedEvent$ = exports.VoiceCallSayEvent$ = exports.VoiceCallReplyEvent$ = exports.VoiceCallPlaybackEvent$ = exports.VoiceCallLanguageEvent$ = exports.VoiceCallInterruptionEvent$ = exports.VoiceCallEndedEvent$ = exports.VoiceCallAgentTranscriptionEvent$ = exports.VoiceAudioFormat$ = exports.Voice$ = exports.TransferWebsocketCommand$ = exports.TransferCommandData$ = exports.TranscriptionWebsocketEvent$ = exports.TranscriptionEvent$ = exports.TextToSpeechOutput$ = exports.TextToSpeechInput$ = void 0;
+exports.TextToSpeech$ = exports.StreamCallEvents$ = exports.SendTransfer$ = exports.SendStop$ = exports.SendSay$ = exports.SendPlay$ = exports.SendHangup$ = exports.ListVoices$ = exports.DescribeVoiceProtocol$ = exports.VoiceCallEvent$ = exports.VoiceSpeechPayload$ = exports.VoiceErrorEvent$ = exports.VoiceConnectParams$ = exports.VoiceCallTranscriptionEvent$ = exports.VoiceCallToolEvent$ = exports.VoiceCallStartedEvent$ = exports.VoiceCallSayEvent$ = exports.VoiceCallReplyEvent$ = exports.VoiceCallPlaybackEvent$ = exports.VoiceCallLanguageEvent$ = exports.VoiceCallInterruptionEvent$ = exports.VoiceCallEndedEvent$ = exports.VoiceCallBackendEvent$ = exports.VoiceCallAgentTranscriptionEvent$ = exports.VoiceAudioFormat$ = exports.Voice$ = exports.TransferWebsocketCommand$ = exports.TransferCommandData$ = exports.TranscriptionWebsocketEvent$ = exports.TranscriptionEvent$ = exports.TextToSpeechOutput$ = exports.TextToSpeechInput$ = void 0;
 const _AE = "AudioEvent";
 const _ATE = "AgentTranscriptionEvent";
 const _ATWE = "AgentTranscriptionWebsocketEvent";
@@ -68,6 +68,7 @@ const _TWE = "TranscriptionWebsocketEvent";
 const _V = "Voice";
 const _VAF = "VoiceAudioFormat";
 const _VCATE = "VoiceCallAgentTranscriptionEvent";
+const _VCBE = "VoiceCallBackendEvent";
 const _VCE = "VoiceCallEvent";
 const _VCEE = "VoiceCallEndedEvent";
 const _VCESB = "VoiceCallEventStreamBody";
@@ -98,6 +99,7 @@ const _aWE = "audioWebsocketEvent";
 const _ac = "action";
 const _as = "as";
 const _au = "audio";
+const _b = "backend";
 const _bS = "backgroundSound";
 const _c = "client";
 const _cC = "callerCompany";
@@ -120,7 +122,9 @@ const _compa = "company";
 const _con = "context";
 const _cr = "credential";
 const _d = "data";
+const _dM = "durationMs";
 const _dN = "dialedNumber";
+const _de = "delivered";
 const _e = "error";
 const _eT = "endTime";
 const _eWE = "errorWebsocketEvent";
@@ -182,7 +186,8 @@ const _sa = "say";
 const _se = "seq";
 const _sp = "speech";
 const _st = "streaming";
-const _sta = "started";
+const _sta = "status";
+const _star = "started";
 const _t = "type";
 const _tCI = "toolCallId";
 const _tI = "traceId";
@@ -193,7 +198,8 @@ const _te = "text";
 const _ti = "time";
 const _to = "to";
 const _too = "tool";
-const _tr = "transcription";
+const _tr = "transport";
+const _tra = "transcription";
 const _u = "uri";
 const _v = "voices";
 const _va = "variables";
@@ -502,6 +508,11 @@ exports.VoiceCallAgentTranscriptionEvent$ = [3, n1, _VCATE,
     [_i, _te, _f, _l, _ti],
     [0, 0, 2, 0, 1], 3
 ];
+exports.VoiceCallBackendEvent$ = [3, n1, _VCBE,
+    0,
+    [_tr, _ev, _de, _sta, _dM, _e],
+    [0, 0, 2, 1, 1, 0], 3
+];
 exports.VoiceCallEndedEvent$ = [3, n1, _VCEE,
     0,
     [_sIe, _r],
@@ -571,8 +582,8 @@ var VoiceList = [1, n1, _VL,
 ];
 exports.VoiceCallEvent$ = [4, n1, _VCE,
     0,
-    [_sta, _tr, _aT, _re, _sa, _pl, _int, _l, _too, _end],
-    [() => exports.VoiceCallStartedEvent$, () => exports.VoiceCallTranscriptionEvent$, () => exports.VoiceCallAgentTranscriptionEvent$, () => exports.VoiceCallReplyEvent$, () => exports.VoiceCallSayEvent$, () => exports.VoiceCallPlaybackEvent$, () => exports.VoiceCallInterruptionEvent$, () => exports.VoiceCallLanguageEvent$, () => exports.VoiceCallToolEvent$, () => exports.VoiceCallEndedEvent$]
+    [_star, _tra, _aT, _re, _sa, _pl, _int, _l, _too, _b, _end],
+    [() => exports.VoiceCallStartedEvent$, () => exports.VoiceCallTranscriptionEvent$, () => exports.VoiceCallAgentTranscriptionEvent$, () => exports.VoiceCallReplyEvent$, () => exports.VoiceCallSayEvent$, () => exports.VoiceCallPlaybackEvent$, () => exports.VoiceCallInterruptionEvent$, () => exports.VoiceCallLanguageEvent$, () => exports.VoiceCallToolEvent$, () => exports.VoiceCallBackendEvent$, () => exports.VoiceCallEndedEvent$]
 ];
 exports.DescribeVoiceProtocol$ = [9, n1, _DVP,
     { [_h]: ["POST", "/v2/voice/describe", 200] }, () => exports.DescribeVoiceProtocolInput$, () => exports.DescribeVoiceProtocolOutput$

package/dist-es/models/enums.js

@@ -20,6 +20,10 @@ export const VoiceControlAction = {
     END_OF_TURN: "end_of_turn",
     KEEPALIVE: "keepalive",
 };
+export const VoiceCallBackendTransport = {
+    SQS: "sqs",
+    WEBHOOK: "webhook",
+};
 export const VoiceCallEndReason = {
     COMPLETE: "complete",
     ERROR: "error",

package/dist-es/schemas/schemas_0.js

@@ -64,6 +64,7 @@ const _TWE = "TranscriptionWebsocketEvent";
 const _V = "Voice";
 const _VAF = "VoiceAudioFormat";
 const _VCATE = "VoiceCallAgentTranscriptionEvent";
+const _VCBE = "VoiceCallBackendEvent";
 const _VCE = "VoiceCallEvent";
 const _VCEE = "VoiceCallEndedEvent";
 const _VCESB = "VoiceCallEventStreamBody";
@@ -94,6 +95,7 @@ const _aWE = "audioWebsocketEvent";
 const _ac = "action";
 const _as = "as";
 const _au = "audio";
+const _b = "backend";
 const _bS = "backgroundSound";
 const _c = "client";
 const _cC = "callerCompany";
@@ -116,7 +118,9 @@ const _compa = "company";
 const _con = "context";
 const _cr = "credential";
 const _d = "data";
+const _dM = "durationMs";
 const _dN = "dialedNumber";
+const _de = "delivered";
 const _e = "error";
 const _eT = "endTime";
 const _eWE = "errorWebsocketEvent";
@@ -178,7 +182,8 @@ const _sa = "say";
 const _se = "seq";
 const _sp = "speech";
 const _st = "streaming";
-const _sta = "started";
+const _sta = "status";
+const _star = "started";
 const _t = "type";
 const _tCI = "toolCallId";
 const _tI = "traceId";
@@ -189,7 +194,8 @@ const _te = "text";
 const _ti = "time";
 const _to = "to";
 const _too = "tool";
-const _tr = "transcription";
+const _tr = "transport";
+const _tra = "transcription";
 const _u = "uri";
 const _v = "voices";
 const _va = "variables";
@@ -498,6 +504,11 @@ export var VoiceCallAgentTranscriptionEvent$ = [3, n1, _VCATE,
     [_i, _te, _f, _l, _ti],
     [0, 0, 2, 0, 1], 3
 ];
+export var VoiceCallBackendEvent$ = [3, n1, _VCBE,
+    0,
+    [_tr, _ev, _de, _sta, _dM, _e],
+    [0, 0, 2, 1, 1, 0], 3
+];
 export var VoiceCallEndedEvent$ = [3, n1, _VCEE,
     0,
     [_sIe, _r],
@@ -567,8 +578,8 @@ var VoiceList = [1, n1, _VL,
 ];
 export var VoiceCallEvent$ = [4, n1, _VCE,
     0,
-    [_sta, _tr, _aT, _re, _sa, _pl, _int, _l, _too, _end],
-    [() => VoiceCallStartedEvent$, () => VoiceCallTranscriptionEvent$, () => VoiceCallAgentTranscriptionEvent$, () => VoiceCallReplyEvent$, () => VoiceCallSayEvent$, () => VoiceCallPlaybackEvent$, () => VoiceCallInterruptionEvent$, () => VoiceCallLanguageEvent$, () => VoiceCallToolEvent$, () => VoiceCallEndedEvent$]
+    [_star, _tra, _aT, _re, _sa, _pl, _int, _l, _too, _b, _end],
+    [() => VoiceCallStartedEvent$, () => VoiceCallTranscriptionEvent$, () => VoiceCallAgentTranscriptionEvent$, () => VoiceCallReplyEvent$, () => VoiceCallSayEvent$, () => VoiceCallPlaybackEvent$, () => VoiceCallInterruptionEvent$, () => VoiceCallLanguageEvent$, () => VoiceCallToolEvent$, () => VoiceCallBackendEvent$, () => VoiceCallEndedEvent$]
 ];
 export var DescribeVoiceProtocol$ = [9, n1, _DVP,
     { [_h]: ["POST", "/v2/voice/describe", 200] }, () => DescribeVoiceProtocolInput$, () => DescribeVoiceProtocolOutput$

package/dist-types/commands/DescribeVoiceProtocolCommand.d.ts

@@ -297,6 +297,14 @@ declare const DescribeVoiceProtocolCommand_base: {
  * //       phase: "started" || "finished", // required
  * //       startMessage: "STRING_VALUE",
  * //     },
+ * //     backend: { // VoiceCallBackendEvent
+ * //       transport: "webhook" || "sqs", // required
+ * //       event: "STRING_VALUE", // required
+ * //       delivered: true || false, // required
+ * //       status: Number("int"),
+ * //       durationMs: Number("int"),
+ * //       error: "STRING_VALUE",
+ * //     },
  * //     ended: { // VoiceCallEndedEvent
  * //       sessionId: "STRING_VALUE", // required
  * //       reason: "hangup" || "transfer" || "timeout" || "error" || "complete", // required

package/dist-types/models/enums.d.ts

@@ -59,6 +59,18 @@ export declare const VoiceControlAction: {
  * @public
  */
 export type VoiceControlAction = (typeof VoiceControlAction)[keyof typeof VoiceControlAction];
+/**
+ * @public
+ * @enum
+ */
+export declare const VoiceCallBackendTransport: {
+    readonly SQS: "sqs";
+    readonly WEBHOOK: "webhook";
+};
+/**
+ * @public
+ */
+export type VoiceCallBackendTransport = (typeof VoiceCallBackendTransport)[keyof typeof VoiceCallBackendTransport];
 /**
  * @public
  * @enum

package/dist-types/models/models_0.d.ts

@@ -1,5 +1,5 @@
 import type { DocumentType as __DocumentType, StreamingBlobTypes } from "@smithy/types";
-import type { VoiceAudioEncoding, VoiceCallEndReason, VoiceCallToolPhase, VoiceCommandType, VoiceControlAction, VoiceEventType, VoiceInputMode, VoicePronunciationEncoding } from "./enums";
+import type { VoiceAudioEncoding, VoiceCallBackendTransport, VoiceCallEndReason, VoiceCallToolPhase, VoiceCommandType, VoiceControlAction, VoiceEventType, VoiceInputMode, VoicePronunciationEncoding } from "./enums";
 /**
  * A pronunciation rule for one phrase. Belongs to a LANGUAGE, not to the agent: 'Wildix' is
  *     respelled differently for an Italian and a German voice, so one list shared ACROSS languages
@@ -182,6 +182,51 @@ export interface VoiceCallAgentTranscriptionEvent {
     language?: string | undefined;
     time?: number | undefined;
 }
+/**
+ * One delivery to the agent's external backend, for a call whose replies this platform does not
+ * produce.
+ *
+ * Without it such a call reads as a call where nothing happens: the transcript shows the greeting
+ * and then silence, and a backend that is unreachable, one that refuses, and one that answers 200
+ * and then says nothing are the same empty panel. Diagnosing that is one of the reasons the
+ * preview exists, so the delivery has to be visible where the call is watched rather than only in
+ * a service log.
+ *
+ * This platform puts no request body, no signature and no credential on it: the envelope holds
+ * the caller's words, and this stream is read by anyone who may watch the call. `error` is the
+ * one member carrying the backend's OWN words, truncated — a backend that echoes what it was
+ * sent echoes it here, which is the tenant's own text reaching the tenant's own operators.
+ * @public
+ */
+export interface VoiceCallBackendEvent {
+    /**
+     * Which transport carries an external backend's events.
+     * @public
+     */
+    transport: VoiceCallBackendTransport;
+    /**
+     * The event delivered, by its name in the backend contract — `started`, `reply`, `playback`, `ended`.
+     * @public
+     */
+    event: string;
+    delivered: boolean;
+    /**
+     * HTTP status for the webhook transport; absent for a queue send, which has none.
+     * @public
+     */
+    status?: number | undefined;
+    /**
+     * How long the delivery took, in milliseconds.
+     * @public
+     */
+    durationMs?: number | undefined;
+    /**
+     * Why the backend did not receive it, truncated. Absent when it did — `delivered` is what says
+     * whether it arrived, and this only ever explains a `false`.
+     * @public
+     */
+    error?: string | undefined;
+}
 /**
  * The session ended. After `transfer` another session of the same call may follow with its own `started`.
  * @public
@@ -356,7 +401,7 @@ export interface VoiceCallTranscriptionEvent {
  * One event on a live call. The SSE frame's `event:` is the member name and `data:` this whole union document, so a frame reads `data: \{"say": \{...\}\}` — never the member's payload on its own.
  * @public
  */
-export type VoiceCallEvent = VoiceCallEvent.AgentTranscriptionMember | VoiceCallEvent.EndedMember | VoiceCallEvent.InterruptionMember | VoiceCallEvent.LanguageMember | VoiceCallEvent.PlaybackMember | VoiceCallEvent.ReplyMember | VoiceCallEvent.SayMember | VoiceCallEvent.StartedMember | VoiceCallEvent.ToolMember | VoiceCallEvent.TranscriptionMember | VoiceCallEvent.$UnknownMember;
+export type VoiceCallEvent = VoiceCallEvent.AgentTranscriptionMember | VoiceCallEvent.BackendMember | VoiceCallEvent.EndedMember | VoiceCallEvent.InterruptionMember | VoiceCallEvent.LanguageMember | VoiceCallEvent.PlaybackMember | VoiceCallEvent.ReplyMember | VoiceCallEvent.SayMember | VoiceCallEvent.StartedMember | VoiceCallEvent.ToolMember | VoiceCallEvent.TranscriptionMember | VoiceCallEvent.$UnknownMember;
 /**
  * @public
  */
@@ -375,6 +420,7 @@ export declare namespace VoiceCallEvent {
         interruption?: never;
         language?: never;
         tool?: never;
+        backend?: never;
         ended?: never;
         $unknown?: never;
     }
@@ -392,6 +438,7 @@ export declare namespace VoiceCallEvent {
         interruption?: never;
         language?: never;
         tool?: never;
+        backend?: never;
         ended?: never;
         $unknown?: never;
     }
@@ -409,6 +456,7 @@ export declare namespace VoiceCallEvent {
         interruption?: never;
         language?: never;
         tool?: never;
+        backend?: never;
         ended?: never;
         $unknown?: never;
     }
@@ -426,6 +474,7 @@ export declare namespace VoiceCallEvent {
         interruption?: never;
         language?: never;
         tool?: never;
+        backend?: never;
         ended?: never;
         $unknown?: never;
     }
@@ -443,6 +492,7 @@ export declare namespace VoiceCallEvent {
         interruption?: never;
         language?: never;
         tool?: never;
+        backend?: never;
         ended?: never;
         $unknown?: never;
     }
@@ -460,6 +510,7 @@ export declare namespace VoiceCallEvent {
         interruption?: never;
         language?: never;
         tool?: never;
+        backend?: never;
         ended?: never;
         $unknown?: never;
     }
@@ -477,6 +528,7 @@ export declare namespace VoiceCallEvent {
         interruption: VoiceCallInterruptionEvent;
         language?: never;
         tool?: never;
+        backend?: never;
         ended?: never;
         $unknown?: never;
     }
@@ -494,6 +546,7 @@ export declare namespace VoiceCallEvent {
         interruption?: never;
         language: VoiceCallLanguageEvent;
         tool?: never;
+        backend?: never;
         ended?: never;
         $unknown?: never;
     }
@@ -511,6 +564,37 @@ export declare namespace VoiceCallEvent {
         interruption?: never;
         language?: never;
         tool: VoiceCallToolEvent;
+        backend?: never;
+        ended?: never;
+        $unknown?: never;
+    }
+    /**
+     * One delivery to the agent's external backend, for a call whose replies this platform does not
+     * produce.
+     *
+     * Without it such a call reads as a call where nothing happens: the transcript shows the greeting
+     * and then silence, and a backend that is unreachable, one that refuses, and one that answers 200
+     * and then says nothing are the same empty panel. Diagnosing that is one of the reasons the
+     * preview exists, so the delivery has to be visible where the call is watched rather than only in
+     * a service log.
+     *
+     * This platform puts no request body, no signature and no credential on it: the envelope holds
+     * the caller's words, and this stream is read by anyone who may watch the call. `error` is the
+     * one member carrying the backend's OWN words, truncated — a backend that echoes what it was
+     * sent echoes it here, which is the tenant's own text reaching the tenant's own operators.
+     * @public
+     */
+    interface BackendMember {
+        started?: never;
+        transcription?: never;
+        agentTranscription?: never;
+        reply?: never;
+        say?: never;
+        playback?: never;
+        interruption?: never;
+        language?: never;
+        tool?: never;
+        backend: VoiceCallBackendEvent;
         ended?: never;
         $unknown?: never;
     }
@@ -528,6 +612,7 @@ export declare namespace VoiceCallEvent {
         interruption?: never;
         language?: never;
         tool?: never;
+        backend?: never;
         ended: VoiceCallEndedEvent;
         $unknown?: never;
     }
@@ -544,6 +629,7 @@ export declare namespace VoiceCallEvent {
         interruption?: never;
         language?: never;
         tool?: never;
+        backend?: never;
         ended?: never;
         $unknown: [string, any];
     }
@@ -561,6 +647,7 @@ export declare namespace VoiceCallEvent {
         interruption: (value: VoiceCallInterruptionEvent) => T;
         language: (value: VoiceCallLanguageEvent) => T;
         tool: (value: VoiceCallToolEvent) => T;
+        backend: (value: VoiceCallBackendEvent) => T;
         ended: (value: VoiceCallEndedEvent) => T;
         _: (name: string, value: any) => T;
     }

package/dist-types/schemas/schemas_0.d.ts

@@ -64,6 +64,7 @@ export declare var TransferWebsocketCommand$: StaticStructureSchema;
 export declare var Voice$: StaticStructureSchema;
 export declare var VoiceAudioFormat$: StaticStructureSchema;
 export declare var VoiceCallAgentTranscriptionEvent$: StaticStructureSchema;
+export declare var VoiceCallBackendEvent$: StaticStructureSchema;
 export declare var VoiceCallEndedEvent$: StaticStructureSchema;
 export declare var VoiceCallInterruptionEvent$: StaticStructureSchema;
 export declare var VoiceCallLanguageEvent$: StaticStructureSchema;

package/package.json

@@ -1,7 +1,7 @@
 {
     "name": "@wildix/wilma-voice-client",
     "description": "@wildix/wilma-voice-client client",
-    "version": "0.0.38",
+    "version": "0.0.39",
     "scripts": {
         "build": "concurrently 'yarn:build:cjs' 'yarn:build:es' 'yarn:build:types'",
         "build:cjs": "tsc -p tsconfig.cjs.json",