@wildix/wilma-voice-client 0.0.41 → 0.0.43

package/dist-cjs/models/enums.js

@@ -1,6 +1,6 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.MediaVadFrameType = exports.MediaStartFrameType = exports.MediaRunPurpose = exports.MediaAudioEncoding = exports.MediaPlaybackFrameType = exports.MediaPlaybackEvent = exports.MediaPlaybackStopCause = exports.MediaFinishFrameType = exports.MediaFinishReason = exports.MediaDtmfFrameType = exports.MediaTransferFrameType = exports.MediaTranscriptionFrameType = exports.MediaParty = exports.MediaStopFrameType = exports.MediaStartedFrameType = exports.MediaRestoreFrameType = exports.MediaLanguageFrameType = exports.MediaHangupFrameType = exports.MediaFinishedFrameType = exports.MediaErrorFrameType = exports.MediaDuckFrameType = exports.MediaAudioEndFrameType = exports.MediaAudioBeginFrameType = exports.VoiceCommandType = exports.VoiceInputMode = exports.VoiceAudioEncoding = exports.VoiceCallToolPhase = exports.VoiceCallEndReason = exports.VoiceControlAction = exports.VoiceEventType = exports.VoicePronunciationEncoding = void 0;
+exports.MediaVadFrameType = exports.MediaStartFrameType = exports.MediaRunPurpose = exports.MediaAudioEncoding = exports.MediaPlaybackFrameType = exports.MediaPlaybackEvent = exports.MediaPlaybackStopCause = exports.MediaFinishFrameType = exports.MediaFinishReason = exports.MediaDtmfFrameType = exports.MediaTransferFrameType = exports.MediaTranscriptionFrameType = exports.MediaParty = exports.MediaStopFrameType = exports.MediaStartedFrameType = exports.MediaRestoreFrameType = exports.MediaLanguageFrameType = exports.MediaHangupFrameType = exports.MediaFinishedFrameType = exports.MediaErrorFrameType = exports.MediaDuckFrameType = exports.MediaAudioEndFrameType = exports.MediaAudioBeginFrameType = exports.VoiceCommandType = exports.VoiceInputMode = exports.VoiceAudioEncoding = exports.VoiceTranscriptionDropReason = exports.VoiceCallToolPhase = exports.VoiceCallEndReason = exports.VoiceControlAction = exports.VoiceEventType = exports.VoicePronunciationEncoding = void 0;
 exports.VoicePronunciationEncoding = {
     ALIAS: "alias",
     IPA: "ipa",
@@ -34,6 +34,10 @@ exports.VoiceCallToolPhase = {
     FINISHED: "finished",
     STARTED: "started",
 };
+exports.VoiceTranscriptionDropReason = {
+    LOW_CONFIDENCE: "low_confidence",
+    OTHER_SPEAKER: "other_speaker",
+};
 exports.VoiceAudioEncoding = {
     ALAW: "alaw",
     MULAW: "mulaw",

package/dist-cjs/schemas/schemas_0.js

@@ -164,6 +164,7 @@ const _dN = "dialedNumber";
 const _dVD = "duckVolumeDb";
 const _di = "digits";
 const _dr = "draft";
+const _dro = "dropped";
 const _dt = "dtmf";
 const _du = "duck";
 const _e = "error";
@@ -205,6 +206,7 @@ const _m = "message";
 const _mC = "minConfidence";
 const _mT = "mediaType";
 const _mW = "minWords";
+const _mWM = "maxWaitMs";
 const _me = "media";
 const _met = "metadata";
 const _n = "name";
@@ -231,9 +233,10 @@ const _rAMe = "restoreAfterMs";
 const _rI = "replyId";
 const _rM = "rampMs";
 const _rWE = "replyWebsocketEvent";
-const _re = "restore";
+const _re = "recogniser";
 const _rep = "reply";
-const _res = "resume";
+const _res = "restore";
+const _resu = "resume";
 const _ret = "retryable";
 const _s = "smithy.ts.sdk.synthetic.wildix.wilma.voice";
 const _sEWE = "sessionEndedWebsocketEvent";
@@ -379,8 +382,8 @@ exports.GetVoiceDefaultsInput$ = [3, n1, _GVDI,
 ];
 exports.GetVoiceDefaultsOutput$ = [3, n1, _GVDO,
     0,
-    [_vD, _in, _tE],
-    [() => exports.VoiceDetectionDefaults$, () => exports.VoiceInterruptionDefaults$, () => exports.VoiceTurnEndDefaults$], 3
+    [_re, _vD, _in, _tE],
+    [0, () => exports.VoiceDetectionDefaults$, () => exports.VoiceInterruptionDefaults$, () => exports.VoiceTurnEndDefaults$], 4
 ];
 exports.HangupCommandData$ = [3, n1, _HCD,
     0,
@@ -429,7 +432,7 @@ exports.MediaAudioFormat$ = [3, n1, _MAF,
 ];
 exports.MediaCloudFrames$ = [3, n1, _MCF,
     0,
-    [_sta, _tr, _aB, _au, _aE, _sto, _du, _re, _l, _tra, _h, _e, _fi],
+    [_sta, _tr, _aB, _au, _aE, _sto, _du, _res, _l, _tra, _h, _e, _fi],
     [() => exports.MediaStartedFrame$, () => exports.MediaTranscriptionFrame$, () => exports.MediaAudioBeginFrame$, 21, () => exports.MediaAudioEndFrame$, () => exports.MediaStopFrame$, () => exports.MediaDuckFrame$, () => exports.MediaRestoreFrame$, () => exports.MediaLanguageFrame$, () => exports.MediaTransferFrame$, () => exports.MediaHangupFrame$, () => exports.MediaErrorFrame$, () => exports.MediaFinishedFrame$]
 ];
 exports.MediaDtmfFrame$ = [3, n1, _MDF,
@@ -509,7 +512,7 @@ exports.MediaStartedFrame$ = [3, n1, _MSF,
 ];
 exports.MediaStartFrame$ = [3, n1, _MSFe,
     0,
-    [_t, _pu, _cI, _fI, _sIe, _res, _rAM, _au, _cal, _ag, _la, _pL, _aD, _dN, _var],
+    [_t, _pu, _cI, _fI, _sIe, _resu, _rAM, _au, _cal, _ag, _la, _pL, _aD, _dN, _var],
     [0, 0, 0, 1, 0, 2, 1, () => exports.MediaAudioFormat$, () => exports.MediaStartCaller$, () => exports.MediaStartAgent$, 64 | 0, 0, 2, 0, 0], 9
 ];
 exports.MediaStopFrame$ = [3, n1, _MSFed,
@@ -759,8 +762,8 @@ exports.VoiceCallToolEvent$ = [3, n1, _VCTE,
 ];
 exports.VoiceCallTranscriptionEvent$ = [3, n1, _VCTEo,
     0,
-    [_i, _te, _f, _l, _ti],
-    [0, 0, 2, 0, 1], 3
+    [_i, _te, _f, _l, _ti, _dro],
+    [0, 0, 2, 0, 1, 0], 3
 ];
 exports.VoiceConnectParams$ = [3, n1, _VCP,
     0,
@@ -789,8 +792,8 @@ exports.VoiceSpeechPayload$ = [3, n1, _VSP,
 ];
 exports.VoiceTurnEndDefaults$ = [3, n1, _VTED,
     0,
-    [_hM],
-    [1], 1
+    [_hM, _mWM],
+    [128 | 1, 1], 2
 ];
 var LanguageCodeList = 64 | 0;
 var VoicePronunciationList = [1, n2, _VPL,
@@ -800,6 +803,7 @@ var MediaStartLanguageList = 64 | 0;
 var VoiceList = [1, n1, _VL,
     0, () => exports.Voice$
 ];
+var VoiceHoldByRecogniser = 128 | 1;
 exports.VoiceCallEvent$ = [4, n1, _VCE,
     0,
     [_sta, _tr, _aT, _rep, _sa, _pl, _inte, _l, _too, _end],

package/dist-es/models/enums.js

@@ -31,6 +31,10 @@ export const VoiceCallToolPhase = {
     FINISHED: "finished",
     STARTED: "started",
 };
+export const VoiceTranscriptionDropReason = {
+    LOW_CONFIDENCE: "low_confidence",
+    OTHER_SPEAKER: "other_speaker",
+};
 export const VoiceAudioEncoding = {
     ALAW: "alaw",
     MULAW: "mulaw",

package/dist-es/schemas/schemas_0.js

@@ -159,6 +159,7 @@ const _dN = "dialedNumber";
 const _dVD = "duckVolumeDb";
 const _di = "digits";
 const _dr = "draft";
+const _dro = "dropped";
 const _dt = "dtmf";
 const _du = "duck";
 const _e = "error";
@@ -200,6 +201,7 @@ const _m = "message";
 const _mC = "minConfidence";
 const _mT = "mediaType";
 const _mW = "minWords";
+const _mWM = "maxWaitMs";
 const _me = "media";
 const _met = "metadata";
 const _n = "name";
@@ -226,9 +228,10 @@ const _rAMe = "restoreAfterMs";
 const _rI = "replyId";
 const _rM = "rampMs";
 const _rWE = "replyWebsocketEvent";
-const _re = "restore";
+const _re = "recogniser";
 const _rep = "reply";
-const _res = "resume";
+const _res = "restore";
+const _resu = "resume";
 const _ret = "retryable";
 const _s = "smithy.ts.sdk.synthetic.wildix.wilma.voice";
 const _sEWE = "sessionEndedWebsocketEvent";
@@ -374,8 +377,8 @@ export var GetVoiceDefaultsInput$ = [3, n1, _GVDI,
 ];
 export var GetVoiceDefaultsOutput$ = [3, n1, _GVDO,
     0,
-    [_vD, _in, _tE],
-    [() => VoiceDetectionDefaults$, () => VoiceInterruptionDefaults$, () => VoiceTurnEndDefaults$], 3
+    [_re, _vD, _in, _tE],
+    [0, () => VoiceDetectionDefaults$, () => VoiceInterruptionDefaults$, () => VoiceTurnEndDefaults$], 4
 ];
 export var HangupCommandData$ = [3, n1, _HCD,
     0,
@@ -424,7 +427,7 @@ export var MediaAudioFormat$ = [3, n1, _MAF,
 ];
 export var MediaCloudFrames$ = [3, n1, _MCF,
     0,
-    [_sta, _tr, _aB, _au, _aE, _sto, _du, _re, _l, _tra, _h, _e, _fi],
+    [_sta, _tr, _aB, _au, _aE, _sto, _du, _res, _l, _tra, _h, _e, _fi],
     [() => MediaStartedFrame$, () => MediaTranscriptionFrame$, () => MediaAudioBeginFrame$, 21, () => MediaAudioEndFrame$, () => MediaStopFrame$, () => MediaDuckFrame$, () => MediaRestoreFrame$, () => MediaLanguageFrame$, () => MediaTransferFrame$, () => MediaHangupFrame$, () => MediaErrorFrame$, () => MediaFinishedFrame$]
 ];
 export var MediaDtmfFrame$ = [3, n1, _MDF,
@@ -504,7 +507,7 @@ export var MediaStartedFrame$ = [3, n1, _MSF,
 ];
 export var MediaStartFrame$ = [3, n1, _MSFe,
     0,
-    [_t, _pu, _cI, _fI, _sIe, _res, _rAM, _au, _cal, _ag, _la, _pL, _aD, _dN, _var],
+    [_t, _pu, _cI, _fI, _sIe, _resu, _rAM, _au, _cal, _ag, _la, _pL, _aD, _dN, _var],
     [0, 0, 0, 1, 0, 2, 1, () => MediaAudioFormat$, () => MediaStartCaller$, () => MediaStartAgent$, 64 | 0, 0, 2, 0, 0], 9
 ];
 export var MediaStopFrame$ = [3, n1, _MSFed,
@@ -754,8 +757,8 @@ export var VoiceCallToolEvent$ = [3, n1, _VCTE,
 ];
 export var VoiceCallTranscriptionEvent$ = [3, n1, _VCTEo,
     0,
-    [_i, _te, _f, _l, _ti],
-    [0, 0, 2, 0, 1], 3
+    [_i, _te, _f, _l, _ti, _dro],
+    [0, 0, 2, 0, 1, 0], 3
 ];
 export var VoiceConnectParams$ = [3, n1, _VCP,
     0,
@@ -784,8 +787,8 @@ export var VoiceSpeechPayload$ = [3, n1, _VSP,
 ];
 export var VoiceTurnEndDefaults$ = [3, n1, _VTED,
     0,
-    [_hM],
-    [1], 1
+    [_hM, _mWM],
+    [128 | 1, 1], 2
 ];
 var LanguageCodeList = 64 | 0;
 var VoicePronunciationList = [1, n2, _VPL,
@@ -795,6 +798,7 @@ var MediaStartLanguageList = 64 | 0;
 var VoiceList = [1, n1, _VL,
     0, () => Voice$
 ];
+var VoiceHoldByRecogniser = 128 | 1;
 export var VoiceCallEvent$ = [4, n1, _VCE,
     0,
     [_sta, _tr, _aT, _rep, _sa, _pl, _inte, _l, _too, _end],

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

@@ -259,6 +259,7 @@ declare const DescribeVoiceProtocolCommand_base: {
  * //       final: true || false, // required
  * //       language: "STRING_VALUE",
  * //       time: Number("long"),
+ * //       dropped: "low_confidence" || "other_speaker",
  * //     },
  * //     agentTranscription: { // VoiceCallAgentTranscriptionEvent
  * //       id: "STRING_VALUE", // required

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

@@ -40,6 +40,7 @@ declare const GetVoiceDefaultsCommand_base: {
  * const command = new GetVoiceDefaultsCommand(input);
  * const response = await client.send(command);
  * // { // GetVoiceDefaultsOutput
+ * //   recogniser: "STRING_VALUE", // required
  * //   voiceDetection: { // VoiceDetectionDefaults
  * //     silenceMs: Number("int"), // required
  * //     silenceLevelDb: Number("int"), // required
@@ -53,7 +54,10 @@ declare const GetVoiceDefaultsCommand_base: {
  * //     restoreAfterMs: Number("int"), // required
  * //   },
  * //   turnEnd: { // VoiceTurnEndDefaults
- * //     holdMs: Number("int"), // required
+ * //     holdMs: { // VoiceHoldByRecogniser // required
+ * //       "<keys>": Number("int"),
+ * //     },
+ * //     maxWaitMs: Number("int"), // required
  * //   },
  * // };
  *

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

@@ -86,6 +86,24 @@ export declare const VoiceCallToolPhase: {
  * @public
  */
 export type VoiceCallToolPhase = (typeof VoiceCallToolPhase)[keyof typeof VoiceCallToolPhase];
+/**
+ * @public
+ * @enum
+ */
+export declare const VoiceTranscriptionDropReason: {
+    /**
+     * The recognition confidence was below the minimum confidence of the call.
+     */
+    readonly LOW_CONFIDENCE: "low_confidence";
+    /**
+     * The recogniser labelled the words with a speaker other than the caller.
+     */
+    readonly OTHER_SPEAKER: "other_speaker";
+};
+/**
+ * @public
+ */
+export type VoiceTranscriptionDropReason = (typeof VoiceTranscriptionDropReason)[keyof typeof VoiceTranscriptionDropReason];
 /**
  * @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 { MediaAudioBeginFrameType, MediaAudioEncoding, MediaAudioEndFrameType, MediaDtmfFrameType, MediaDuckFrameType, MediaErrorFrameType, MediaFinishedFrameType, MediaFinishFrameType, MediaFinishReason, MediaHangupFrameType, MediaLanguageFrameType, MediaParty, MediaPlaybackEvent, MediaPlaybackFrameType, MediaPlaybackStopCause, MediaRestoreFrameType, MediaRunPurpose, MediaStartedFrameType, MediaStartFrameType, MediaStopFrameType, MediaTranscriptionFrameType, MediaTransferFrameType, MediaVadFrameType, VoiceAudioEncoding, VoiceCallEndReason, VoiceCallToolPhase, VoiceCommandType, VoiceControlAction, VoiceEventType, VoiceInputMode, VoicePronunciationEncoding } from "./enums";
+import type { MediaAudioBeginFrameType, MediaAudioEncoding, MediaAudioEndFrameType, MediaDtmfFrameType, MediaDuckFrameType, MediaErrorFrameType, MediaFinishedFrameType, MediaFinishFrameType, MediaFinishReason, MediaHangupFrameType, MediaLanguageFrameType, MediaParty, MediaPlaybackEvent, MediaPlaybackFrameType, MediaPlaybackStopCause, MediaRestoreFrameType, MediaRunPurpose, MediaStartedFrameType, MediaStartFrameType, MediaStopFrameType, MediaTranscriptionFrameType, MediaTransferFrameType, MediaVadFrameType, VoiceAudioEncoding, VoiceCallEndReason, VoiceCallToolPhase, VoiceCommandType, VoiceControlAction, VoiceEventType, VoiceInputMode, VoicePronunciationEncoding, VoiceTranscriptionDropReason } 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
@@ -351,6 +351,11 @@ export interface VoiceCallTranscriptionEvent {
      * @public
      */
     time?: number | undefined;
+    /**
+     * Present on a final that did not go to the agent as a caller turn. It came while the agent spoke and the cloud took it for noise or echo, or for another speaker.
+     * @public
+     */
+    dropped?: VoiceTranscriptionDropReason | undefined;
 }
 /**
  * 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.
@@ -1901,10 +1906,15 @@ export interface VoiceInterruptionDefaults {
  */
 export interface VoiceTurnEndDefaults {
     /**
-     * Wait in ms after the recogniser marks the end of a caller phrase, before the agent answers.
+     * Default of `holdMs` for each recogniser, by the URI scheme of the recogniser, for example `soniox` and `aws`.
      * @public
      */
-    holdMs: number;
+    holdMs: Record<string, number>;
+    /**
+     * Default of `maxWaitMs`.
+     * @public
+     */
+    maxWaitMs: number;
 }
 /**
  * Defaults of `VoiceChannelSettings.voiceDetection`.
@@ -1931,6 +1941,11 @@ export interface VoiceDetectionDefaults {
  * @public
  */
 export interface GetVoiceDefaultsOutput {
+    /**
+     * URI scheme of the recogniser that a call uses when its agent sets none, for example `soniox`.
+     * @public
+     */
+    recogniser: string;
     /**
      * Defaults of `VoiceChannelSettings.voiceDetection`.
      * @public

package/package.json

@@ -1,7 +1,7 @@
 {
     "name": "@wildix/wilma-voice-client",
     "description": "@wildix/wilma-voice-client client",
-    "version": "0.0.41",
+    "version": "0.0.43",
     "scripts": {
         "build": "concurrently 'yarn:build:cjs' 'yarn:build:es' 'yarn:build:types'",
         "build:cjs": "tsc -p tsconfig.cjs.json",