@wildix/wilma-voice-client 0.0.50 → 0.0.52

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

@@ -40,8 +40,7 @@ declare const SendSayCommand_base: {
  *   text: "STRING_VALUE", // required
  *   replyId: "STRING_VALUE", // required
  *   interruptible: true || false,
- *   voice: "STRING_VALUE",
- *   audio: { // SayCommandAudioOptions
+ *   audio: { // VoiceSayCommandAudioOptions
  *     streamId: "STRING_VALUE", // required
  *     sampleRate: Number("int"), // required
  *   },

package/dist-types/index.d.ts

@@ -5,12 +5,11 @@
  *     (engine, instructions, capabilities and channels.voice settings). This is a runtime only: there
  *     is no agent/bot CRUD here — agents are authored and versioned in wilma-agents.
  *
- *     On the WebSocket (transport host `wilma-voice-ws.wildix.com`), PBX clients transcribe caller
- *     speech themselves and send `transcription` / `reply` text, while web/app clients connect with
- *     inputMode=AUDIO and push raw caller `audio` chunks that the service transcribes server-side;
- *     the service returns `say` / `play` / `stop` / `hangup` / `transfer`. On the PBX media socket
- *     (`/v2/voice/media/ws`, subprotocol `wilma-media-v2`), the PBX sends raw caller audio and the
- *     service answers with audio and the text frames listed under DescribeVoiceProtocol `media`.
+ *     On the WebSocket (transport host `wilma-voice-ws.wildix.com`), the PBX transcribes caller
+ *     speech itself and sends `transcription` / `reply` text; the service returns `say` / `play` /
+ *     `stop` / `hangup` / `transfer`. On the call socket
+ *     (`/v2/voice/calls/ws`), the PBX sends raw caller audio and the
+ *     service answers with audio and the text frames listed under DescribeVoiceProtocol `callSocket`.
  *     Every WebSocket message shape (connect params, events and commands) is declared as a structure
  *     and enumerated by DescribeVoiceProtocol so the generated SDKs carry the types and
  *     (de)serializers; the WebSocket transport itself is hand-coded. The HTTP Send* operations push

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

@@ -22,15 +22,15 @@ export type VoicePronunciationEncoding = (typeof VoicePronunciationEncoding)[key
  */
 export declare const VoiceEventType: {
     readonly AGENT_TRANSCRIPTION: "agentTranscription";
-    readonly AUDIO: "audio";
     readonly COMPLETE: "complete";
-    readonly CONTROL: "control";
     readonly ENDED: "ended";
     readonly ERROR: "error";
     readonly INTERRUPTION: "interruption";
     readonly PLAYBACK: "playback";
     readonly REPLY: "reply";
     readonly STARTED: "started";
+    readonly THINKING: "thinking";
+    readonly TOOL: "tool";
     readonly TRANSCRIPTION: "transcription";
 };
 /**
@@ -41,62 +41,40 @@ export type VoiceEventType = (typeof VoiceEventType)[keyof typeof VoiceEventType
  * @public
  * @enum
  */
-export declare const VoiceControlAction: {
-    /**
-     * Discard buffered caller audio and cancel the in-progress input turn.
-     */
-    readonly CLEAR: "clear";
-    /**
-     * Push-to-talk: the caller finished the current utterance; transcribe and run the agent now.
-     */
-    readonly END_OF_TURN: "end_of_turn";
-    /**
-     * Application-level keepalive.
-     */
-    readonly KEEPALIVE: "keepalive";
-};
-/**
- * @public
- */
-export type VoiceControlAction = (typeof VoiceControlAction)[keyof typeof VoiceControlAction];
-/**
- * @public
- * @enum
- */
-export declare const VoiceCallEndReason: {
-    readonly COMPLETE: "complete";
-    readonly ERROR: "error";
+export declare const VoiceCommandType: {
     readonly HANGUP: "hangup";
-    readonly TIMEOUT: "timeout";
+    readonly PLAY: "play";
+    readonly SAY: "say";
+    readonly STOP: "stop";
     readonly TRANSFER: "transfer";
 };
 /**
  * @public
  */
-export type VoiceCallEndReason = (typeof VoiceCallEndReason)[keyof typeof VoiceCallEndReason];
+export type VoiceCommandType = (typeof VoiceCommandType)[keyof typeof VoiceCommandType];
 /**
  * @public
  * @enum
  */
-export declare const VoiceCallToolPhase: {
+export declare const VoiceToolPhase: {
     readonly FINISHED: "finished";
     readonly STARTED: "started";
 };
 /**
  * @public
  */
-export type VoiceCallToolPhase = (typeof VoiceCallToolPhase)[keyof typeof VoiceCallToolPhase];
+export type VoiceToolPhase = (typeof VoiceToolPhase)[keyof typeof VoiceToolPhase];
 /**
  * @public
  * @enum
  */
 export declare const VoiceTranscriptionDropReason: {
     /**
-     * The recognition confidence was below the minimum confidence of the call.
+     * The confidence of the recogniser was less than the minimum confidence of the call.
      */
     readonly LOW_CONFIDENCE: "low_confidence";
     /**
-     * The recogniser labelled the words with a speaker other than the caller.
+     * The recogniser identified a speaker that is not the caller.
      */
     readonly OTHER_SPEAKER: "other_speaker";
 };
@@ -108,221 +86,58 @@ export type VoiceTranscriptionDropReason = (typeof VoiceTranscriptionDropReason)
  * @public
  * @enum
  */
-export declare const VoiceAudioEncoding: {
-    /**
-     * G.711 A-law.
-     */
-    readonly ALAW: "alaw";
-    /**
-     * G.711 μ-law.
-     */
-    readonly MULAW: "mulaw";
-    /**
-     * Opus.
-     */
-    readonly OPUS: "opus";
-    /**
-     * 16-bit signed little-endian linear PCM.
-     */
-    readonly PCM_S16LE: "pcm_s16le";
-};
-/**
- * @public
- */
-export type VoiceAudioEncoding = (typeof VoiceAudioEncoding)[keyof typeof VoiceAudioEncoding];
-/**
- * @public
- * @enum
- */
-export declare const VoiceInputMode: {
-    /**
-     * The client pushes raw caller `audio` chunks and the service transcribes them server-side (web/app clients).
-     */
-    readonly AUDIO: "audio";
-    /**
-     * The client sends `transcription` / `reply` text events; it performs speech-to-text itself (PBX behaviour).
-     */
-    readonly TRANSCRIPTION: "transcription";
-};
-/**
- * @public
- */
-export type VoiceInputMode = (typeof VoiceInputMode)[keyof typeof VoiceInputMode];
-/**
- * @public
- * @enum
- */
-export declare const VoiceCommandType: {
+export declare const VoiceCallEndReason: {
+    readonly COMPLETE: "complete";
+    readonly ERROR: "error";
     readonly HANGUP: "hangup";
-    readonly PLAY: "play";
-    readonly SAY: "say";
-    readonly STOP: "stop";
+    readonly TIMEOUT: "timeout";
     readonly TRANSFER: "transfer";
 };
 /**
  * @public
  */
-export type VoiceCommandType = (typeof VoiceCommandType)[keyof typeof VoiceCommandType];
+export type VoiceCallEndReason = (typeof VoiceCallEndReason)[keyof typeof VoiceCallEndReason];
 /**
  * @public
  * @enum
  */
-export declare const MediaAudioBeginFrameType: {
+export declare const VoiceCallServerFrameType: {
     readonly AUDIO_BEGIN: "audio.begin";
-};
-/**
- * @public
- */
-export type MediaAudioBeginFrameType = (typeof MediaAudioBeginFrameType)[keyof typeof MediaAudioBeginFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaAudioEndFrameType: {
     readonly AUDIO_END: "audio.end";
-};
-/**
- * @public
- */
-export type MediaAudioEndFrameType = (typeof MediaAudioEndFrameType)[keyof typeof MediaAudioEndFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaDuckFrameType: {
     readonly DUCK: "duck";
-};
-/**
- * @public
- */
-export type MediaDuckFrameType = (typeof MediaDuckFrameType)[keyof typeof MediaDuckFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaErrorFrameType: {
+    readonly ENDED: "ended";
     readonly ERROR: "error";
-};
-/**
- * @public
- */
-export type MediaErrorFrameType = (typeof MediaErrorFrameType)[keyof typeof MediaErrorFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaFinishedFrameType: {
-    readonly FINISHED: "finished";
-};
-/**
- * @public
- */
-export type MediaFinishedFrameType = (typeof MediaFinishedFrameType)[keyof typeof MediaFinishedFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaHangupFrameType: {
     readonly HANGUP: "hangup";
-};
-/**
- * @public
- */
-export type MediaHangupFrameType = (typeof MediaHangupFrameType)[keyof typeof MediaHangupFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaLanguageFrameType: {
     readonly LANGUAGE: "language";
-};
-/**
- * @public
- */
-export type MediaLanguageFrameType = (typeof MediaLanguageFrameType)[keyof typeof MediaLanguageFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaRestoreFrameType: {
-    readonly RESTORE: "restore";
-};
-/**
- * @public
- */
-export type MediaRestoreFrameType = (typeof MediaRestoreFrameType)[keyof typeof MediaRestoreFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaStartedFrameType: {
     readonly STARTED: "started";
-};
-/**
- * @public
- */
-export type MediaStartedFrameType = (typeof MediaStartedFrameType)[keyof typeof MediaStartedFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaStopFrameType: {
     readonly STOP: "stop";
-};
-/**
- * @public
- */
-export type MediaStopFrameType = (typeof MediaStopFrameType)[keyof typeof MediaStopFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaParty: {
-    readonly CALLEE: "callee";
-    readonly CALLER: "caller";
-};
-/**
- * @public
- */
-export type MediaParty = (typeof MediaParty)[keyof typeof MediaParty];
-/**
- * @public
- * @enum
- */
-export declare const MediaTranscriptionFrameType: {
     readonly TRANSCRIPTION: "transcription";
-};
-/**
- * @public
- */
-export type MediaTranscriptionFrameType = (typeof MediaTranscriptionFrameType)[keyof typeof MediaTranscriptionFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaTransferFrameType: {
     readonly TRANSFER: "transfer";
 };
 /**
  * @public
  */
-export type MediaTransferFrameType = (typeof MediaTransferFrameType)[keyof typeof MediaTransferFrameType];
+export type VoiceCallServerFrameType = (typeof VoiceCallServerFrameType)[keyof typeof VoiceCallServerFrameType];
 /**
  * @public
  * @enum
  */
-export declare const MediaDtmfFrameType: {
+export declare const VoiceCallClientFrameType: {
     readonly DTMF: "dtmf";
+    readonly END: "end";
+    readonly PLAYBACK: "playback";
+    readonly START: "start";
+    readonly VAD: "vad";
 };
 /**
  * @public
  */
-export type MediaDtmfFrameType = (typeof MediaDtmfFrameType)[keyof typeof MediaDtmfFrameType];
+export type VoiceCallClientFrameType = (typeof VoiceCallClientFrameType)[keyof typeof VoiceCallClientFrameType];
 /**
  * @public
  * @enum
  */
-export declare const MediaFinishReason: {
+export declare const VoiceCallClientEndReason: {
     /**
      * The PBX ran the agent's `hangup`.
      */
@@ -331,10 +146,6 @@ export declare const MediaFinishReason: {
      * The PBX ran the agent's `transfer`, or tried to and the target does not exist.
      */
     readonly AGENT_TRANSFER: "agent-transfer";
-    /**
-     * The callee hung up.
-     */
-    readonly CALLEE_HANGUP: "callee-hangup";
     /**
      * The caller hung up.
      */
@@ -347,45 +158,12 @@ export declare const MediaFinishReason: {
 /**
  * @public
  */
-export type MediaFinishReason = (typeof MediaFinishReason)[keyof typeof MediaFinishReason];
+export type VoiceCallClientEndReason = (typeof VoiceCallClientEndReason)[keyof typeof VoiceCallClientEndReason];
 /**
  * @public
  * @enum
  */
-export declare const MediaFinishFrameType: {
-    readonly FINISH: "finish";
-};
-/**
- * @public
- */
-export type MediaFinishFrameType = (typeof MediaFinishFrameType)[keyof typeof MediaFinishFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaPlaybackStopCause: {
-    /**
-     * The caller hung up or the run ended.
-     */
-    readonly CALLER: "caller";
-    /**
-     * A `stop` frame from the cloud.
-     */
-    readonly CLOUD: "cloud";
-    /**
-     * The PBX dropped the stream, for example at its queue limit.
-     */
-    readonly DROP: "drop";
-};
-/**
- * @public
- */
-export type MediaPlaybackStopCause = (typeof MediaPlaybackStopCause)[keyof typeof MediaPlaybackStopCause];
-/**
- * @public
- * @enum
- */
-export declare const MediaPlaybackEvent: {
+export declare const VoiceCallPlaybackPhase: {
     /**
      * The stream played to its end.
      */
@@ -402,66 +180,30 @@ export declare const MediaPlaybackEvent: {
 /**
  * @public
  */
-export type MediaPlaybackEvent = (typeof MediaPlaybackEvent)[keyof typeof MediaPlaybackEvent];
-/**
- * @public
- * @enum
- */
-export declare const MediaPlaybackFrameType: {
-    readonly PLAYBACK: "playback";
-};
-/**
- * @public
- */
-export type MediaPlaybackFrameType = (typeof MediaPlaybackFrameType)[keyof typeof MediaPlaybackFrameType];
+export type VoiceCallPlaybackPhase = (typeof VoiceCallPlaybackPhase)[keyof typeof VoiceCallPlaybackPhase];
 /**
  * @public
  * @enum
  */
-export declare const MediaAudioEncoding: {
-    readonly PCM_S16LE: "pcm_s16le";
+export declare const VoiceCallParty: {
+    readonly CALLEE: "callee";
+    readonly CALLER: "caller";
 };
 /**
  * @public
  */
-export type MediaAudioEncoding = (typeof MediaAudioEncoding)[keyof typeof MediaAudioEncoding];
+export type VoiceCallParty = (typeof VoiceCallParty)[keyof typeof VoiceCallParty];
 /**
  * @public
  * @enum
  */
-export declare const MediaRunPurpose: {
-    /**
-     * A voice agent answers the call.
-     */
-    readonly AGENT: "agent";
+export declare const VoiceInputMode: {
     /**
-     * Reserved: the cloud transcribes the call and sends no audio. The cloud does not serve it at this time and closes the socket with 4400.
+     * The client does speech-to-text itself and sends `transcription` / `reply` text events.
      */
     readonly TRANSCRIPTION: "transcription";
 };
 /**
  * @public
  */
-export type MediaRunPurpose = (typeof MediaRunPurpose)[keyof typeof MediaRunPurpose];
-/**
- * @public
- * @enum
- */
-export declare const MediaStartFrameType: {
-    readonly START: "start";
-};
-/**
- * @public
- */
-export type MediaStartFrameType = (typeof MediaStartFrameType)[keyof typeof MediaStartFrameType];
-/**
- * @public
- * @enum
- */
-export declare const MediaVadFrameType: {
-    readonly VAD: "vad";
-};
-/**
- * @public
- */
-export type MediaVadFrameType = (typeof MediaVadFrameType)[keyof typeof MediaVadFrameType];
+export type VoiceInputMode = (typeof VoiceInputMode)[keyof typeof VoiceInputMode];