telnyx 6.71.0 → 6.72.0

package/src/client.ts

@@ -819,11 +819,6 @@ import {
   SiprecConnectorUpdateResponse,
   SiprecConnectors,
 } from './resources/siprec-connectors';
-import {
-  SpeechToText,
-  SpeechToTextListProvidersParams,
-  SpeechToTextListProvidersResponse,
-} from './resources/speech-to-text';
 import {
   SubNumberOrder,
   SubNumberOrderCancelResponse,
@@ -1536,6 +1531,13 @@ import {
   SimCardUpdateResponse,
   SimCards,
 } from './resources/sim-cards/sim-cards';
+import {
+  SpeechToText,
+  SpeechToTextListProvidersParams,
+  SpeechToTextListProvidersResponse,
+  TranscribeClientEvent,
+  TranscribeServerEvent,
+} from './resources/speech-to-text/speech-to-text';
 import { Storage, StorageListMigrationSourceCoverageResponse } from './resources/storage/storage';
 import { TermsOfService } from './resources/terms-of-service/terms-of-service';
 import {
@@ -1549,8 +1551,9 @@ import {
   StreamClientEvent,
   StreamServerEvent,
   TextToSpeech,
-  TextToSpeechGenerateParams,
-  TextToSpeechGenerateResponse,
+  TextToSpeechCreateSpeechParams,
+  TextToSpeechCreateSpeechResponse,
+  TextToSpeechGenerateSpeechParams,
   TextToSpeechListVoicesParams,
   TextToSpeechListVoicesResponse,
 } from './resources/text-to-speech/text-to-speech';
@@ -2791,6 +2794,10 @@ export class Telnyx {
    */
   regulatoryRequirements: API.RegulatoryRequirements = new API.RegulatoryRequirements(this);
   reports: API.Reports = new API.Reports(this);
+  /**
+   * Discover available speech-to-text providers, models, and supported languages.
+   */
+  speechToText: API.SpeechToText = new API.SpeechToText(this);
   /**
    * Requirement Groups
    */
@@ -2984,10 +2991,6 @@ export class Telnyx {
    * UAC connection operations
    */
   uacConnections: API.UacConnections = new API.UacConnections(this);
-  /**
-   * Discover available speech-to-text providers, models, and supported languages.
-   */
-  speechToText: API.SpeechToText = new API.SpeechToText(this);
   /**
    * Retrieve raw Voice SDK call report stats payloads for WebRTC call troubleshooting.
    */
@@ -3102,6 +3105,7 @@ Telnyx.Recordings = Recordings;
 Telnyx.Regions = Regions;
 Telnyx.RegulatoryRequirements = RegulatoryRequirements;
 Telnyx.Reports = Reports;
+Telnyx.SpeechToText = SpeechToText;
 Telnyx.RequirementGroups = RequirementGroups;
 Telnyx.RequirementTypes = RequirementTypes;
 Telnyx.Requirements = Requirements;
@@ -3158,7 +3162,6 @@ Telnyx.Reputation = Reputation;
 Telnyx.TermsOfService = TermsOfService;
 Telnyx.PronunciationDicts = PronunciationDicts;
 Telnyx.UacConnections = UacConnections;
-Telnyx.SpeechToText = SpeechToText;
 Telnyx.VoiceSDKCallReports = VoiceSDKCallReports;
 
 export declare namespace Telnyx {
@@ -4397,6 +4400,14 @@ export declare namespace Telnyx {
     type ReportListWdrsParams as ReportListWdrsParams,
   };
 
+  export {
+    SpeechToText as SpeechToText,
+    type SpeechToTextListProvidersResponse as SpeechToTextListProvidersResponse,
+    type TranscribeClientEvent as TranscribeClientEvent,
+    type TranscribeServerEvent as TranscribeServerEvent,
+    type SpeechToTextListProvidersParams as SpeechToTextListProvidersParams,
+  };
+
   export {
     RequirementGroups as RequirementGroups,
     type RequirementGroup as RequirementGroup,
@@ -4615,11 +4626,12 @@ export declare namespace Telnyx {
 
   export {
     TextToSpeech as TextToSpeech,
-    type TextToSpeechGenerateResponse as TextToSpeechGenerateResponse,
+    type TextToSpeechCreateSpeechResponse as TextToSpeechCreateSpeechResponse,
     type TextToSpeechListVoicesResponse as TextToSpeechListVoicesResponse,
     type StreamClientEvent as StreamClientEvent,
     type StreamServerEvent as StreamServerEvent,
-    type TextToSpeechGenerateParams as TextToSpeechGenerateParams,
+    type TextToSpeechCreateSpeechParams as TextToSpeechCreateSpeechParams,
+    type TextToSpeechGenerateSpeechParams as TextToSpeechGenerateSpeechParams,
     type TextToSpeechListVoicesParams as TextToSpeechListVoicesParams,
   };
 
@@ -4942,12 +4954,6 @@ export declare namespace Telnyx {
     type UacConnectionListParams as UacConnectionListParams,
   };
 
-  export {
-    SpeechToText as SpeechToText,
-    type SpeechToTextListProvidersResponse as SpeechToTextListProvidersResponse,
-    type SpeechToTextListProvidersParams as SpeechToTextListProvidersParams,
-  };
-
   export {
     VoiceSDKCallReports as VoiceSDKCallReports,
     type VoiceSDKCallReportLogEntry as VoiceSDKCallReportLogEntry,

package/src/resources/ai/ai.ts

@@ -180,8 +180,8 @@ export class AI extends APIResource {
     params: AICreateResponseParams,
     options?: RequestOptions,
   ): APIPromise<AICreateResponseResponse> {
-    const { body } = params;
-    return this._client.post('/ai/responses', { body: body, ...options });
+    const { input } = params;
+    return this._client.post('/ai/responses', { body: input, ...options });
   }
 
   /**
@@ -374,7 +374,7 @@ export namespace AISummarizeResponse {
 }
 
 export interface AICreateResponseParams {
-  body: { [key: string]: unknown };
+  input: { [key: string]: unknown };
 }
 
 export interface AISummarizeParams {

package/src/resources/index.ts

@@ -1174,6 +1174,8 @@ export {
 export {
   SpeechToText,
   type SpeechToTextListProvidersResponse,
+  type TranscribeClientEvent,
+  type TranscribeServerEvent,
   type SpeechToTextListProvidersParams,
 } from './speech-to-text';
 export { Storage, type StorageListMigrationSourceCoverageResponse } from './storage/storage';
@@ -1234,11 +1236,12 @@ export {
 } from './texml-applications';
 export {
   TextToSpeech,
-  type TextToSpeechGenerateResponse,
+  type TextToSpeechCreateSpeechResponse,
   type TextToSpeechListVoicesResponse,
   type StreamClientEvent,
   type StreamServerEvent,
-  type TextToSpeechGenerateParams,
+  type TextToSpeechCreateSpeechParams,
+  type TextToSpeechGenerateSpeechParams,
   type TextToSpeechListVoicesParams,
   type TextToSpeechStreamParams,
 } from './text-to-speech';

package/src/resources/speech-to-text/index.ts

@@ -2,13 +2,10 @@
 
 export {
   SpeechToText,
-  type SpeechToTextTranscribeParams,
-  type SpeechToTextStreamParams,
-  type SttClientEvent,
-  type SttServerEvent,
-  type TranscriptFrame,
-  type SttErrorFrame,
+  type SpeechToTextListProvidersResponse,
+  type TranscribeClientEvent,
+  type TranscribeServerEvent,
+  type SpeechToTextListProvidersParams,
 } from './speech-to-text';
-
-export { SpeechToTextWS } from './ws';
-export { WebSocketError } from './internal-base';
+export { type SpeechToTextWSClientOptions } from './ws';
+export { type SpeechToTextWSReconnectOptions } from './ws-base';

package/src/resources/speech-to-text/internal-base.ts

@@ -2,18 +2,32 @@
 
 import * as SpeechToTextAPI from './speech-to-text';
 import { Telnyx } from '../../client';
-
 import { EventEmitter } from '../../core/EventEmitter';
 import { TelnyxError } from '../../core/error';
-import { stringifyQuery } from '../../internal/utils';
+
+import type { RawWebSocketData, ReconnectingEvent, UnsentMessage } from '../../internal/ws';
+
+export type SpeechToTextStreamMessage =
+  | { type: 'connecting' | 'open' | 'closing' }
+  | {
+      type: 'close';
+      code: number;
+      reason: string;
+      unsent: UnsentMessage<SpeechToTextAPI.TranscribeClientEvent>[];
+    }
+  | { type: 'reconnecting'; reconnect: ReconnectingEvent }
+  | { type: 'reconnected' }
+  | { type: 'message'; message: SpeechToTextAPI.TranscribeServerEvent }
+  | { type: 'raw'; data: RawWebSocketData }
+  | { type: 'error'; error: WebSocketError };
 
 export class WebSocketError extends TelnyxError {
   /**
    * The error data that the API sent back in an error event.
    */
-  error?: SpeechToTextAPI.SttServerEvent.ErrorFrame | undefined;
+  error?: SpeechToTextAPI.TranscribeServerEvent.SttErrorFrame | undefined;
 
-  constructor(message: string, event: SpeechToTextAPI.SttServerEvent.ErrorFrame | null) {
+  constructor(message: string, event: SpeechToTextAPI.TranscribeServerEvent.SttErrorFrame | null) {
     super(message);
 
     this.error = event ?? undefined;
@@ -22,32 +36,48 @@ export class WebSocketError extends TelnyxError {
 
 type Simplify<T> = { [KeyType in keyof T]: T[KeyType] } & {};
 
-type WebsocketEvents = Simplify<
+type WebSocketEvents = Simplify<
   {
-    event: (event: SpeechToTextAPI.SttServerEvent) => void;
+    event: (event: SpeechToTextAPI.TranscribeServerEvent) => void;
+    raw: (data: RawWebSocketData) => void;
     error: (error: WebSocketError) => void;
+    close: (
+      code: number,
+      reason: string,
+      unsent: UnsentMessage<SpeechToTextAPI.TranscribeClientEvent>[],
+    ) => void;
+    reconnecting: (event: ReconnectingEvent) => void;
+    reconnected: () => void;
   } & {
-    [EventType in Exclude<NonNullable<SpeechToTextAPI.SttServerEvent['type']>, 'error'>]: (
-      event: Extract<SpeechToTextAPI.SttServerEvent, { type?: EventType }>,
+    [EventType in Exclude<NonNullable<SpeechToTextAPI.TranscribeServerEvent['type']>, 'error'>]: (
+      event: Extract<SpeechToTextAPI.TranscribeServerEvent, { type?: EventType }>,
     ) => unknown;
   }
 >;
 
-export abstract class SpeechToTextEmitter extends EventEmitter<WebsocketEvents> {
+export abstract class SpeechToTextEmitter extends EventEmitter<WebSocketEvents> {
+  /**
+   * Send an event to the API.
+   */
+  abstract send(event: SpeechToTextAPI.TranscribeClientEvent): void;
+
   /**
-   * Send binary audio data to the API.
+   * Send raw data over the WebSocket without JSON serialization.
    */
-  abstract send(data: SpeechToTextAPI.SttClientEvent): void;
+  abstract sendRaw(data: RawWebSocketData): void;
 
   /**
-   * Close the websocket connection.
+   * Close the WebSocket connection.
    */
   abstract close(props?: { code: number; reason: string }): void;
 
   protected _onError(event: null, message: string, cause: any): void;
-  protected _onError(event: SpeechToTextAPI.SttServerEvent.ErrorFrame, message?: string | undefined): void;
   protected _onError(
-    event: SpeechToTextAPI.SttServerEvent.ErrorFrame | null,
+    event: SpeechToTextAPI.TranscribeServerEvent.SttErrorFrame,
+    message?: string | undefined,
+  ): void;
+  protected _onError(
+    event: SpeechToTextAPI.TranscribeServerEvent.SttErrorFrame | null,
     message?: string | undefined,
     cause?: any,
   ): void {
@@ -73,14 +103,11 @@ export abstract class SpeechToTextEmitter extends EventEmitter<WebsocketEvents>
   }
 }
 
-export function buildURL(client: Telnyx, query?: SpeechToTextAPI.SpeechToTextStreamParams | null): URL {
-  const path = '/speech-to-text/transcription';
-  const baseURL = client.baseURL;
-  const url = new URL(baseURL + (baseURL.endsWith('/') ? path.slice(1) : path));
-  if (query) {
-    url.search = stringifyQuery(query);
-  }
-  url.protocol = 'wss';
+export function buildURL(client: Telnyx, parameters: Record<string, unknown>): URL {
+  const { ...query } = parameters;
+  const endpoint = '/speech-to-text/transcription';
+  const url = new URL(client.buildURL(endpoint, query, undefined));
+  url.protocol = url.protocol === 'http:' || url.protocol === 'ws:' ? 'ws:' : 'wss:';
   return url;
 }
 

package/src/resources/speech-to-text/speech-to-text.ts

@@ -2,198 +2,172 @@
 
 import { APIResource } from '../../core/resource';
 import { APIPromise } from '../../core/api-promise';
-import { buildHeaders } from '../../internal/headers';
+import { type Uploadable } from '../../core/uploads';
 import { RequestOptions } from '../../internal/request-options';
 
 /**
- * Speech to text streaming operations via WebSocket
+ * Discover available speech-to-text providers, models, and supported languages.
  */
 export class SpeechToText extends APIResource {
   /**
-   * Open a WebSocket connection to stream audio and receive transcriptions in
-   * real-time. Authentication is provided via the standard
-   * `Authorization: Bearer <API_KEY>` header.
+   * Retrieve the canonical list of supported speech-to-text providers, models,
+   * accepted language codes, and the service types each model supports.
    *
-   * Supported engines: `Azure`, `Deepgram`, `Google`, `Telnyx`.
+   * Service types:
    *
-   * **Connection flow:**
+   * - `streaming` — standalone WebSocket transcription via
+   *   `/speech-to-text/transcription`.
+   * - `file_transcription` — file-based transcription via
+   *   `/ai/audio/transcriptions`.
+   * - `in_call_transcription` — live call transcription via Call Control
+   *   `transcription_start`.
    *
-   * 1. Open WebSocket with query parameters specifying engine, input format, and
-   *    language.
-   * 2. Send binary audio frames (mp3/wav format).
-   * 3. Receive JSON transcript frames with `transcript`, `is_final`, and
-   *    `confidence` fields.
-   * 4. Close connection when done.
-   */
-  transcribe(query: SpeechToTextTranscribeParams, options?: RequestOptions): APIPromise<void> {
-    return this._client.get('/speech-to-text/transcription', {
+   * Use this endpoint to discover which (provider, model) combinations are available
+   * for the surface you need, and which language codes each accepts. `auto` in a
+   * `languages` array indicates the provider performs language detection.
+   */
+  listProviders(
+    query: SpeechToTextListProvidersParams | null | undefined = {},
+    options?: RequestOptions,
+  ): APIPromise<SpeechToTextListProvidersResponse> {
+    return this._client.get('/speech-to-text/providers', {
       query,
+      defaultBaseURL: 'https://api.telnyx.com/v2',
       ...options,
-      headers: buildHeaders([
-        { 'Content-Type': 'application/octet-stream', Accept: '*/*' },
-        options?.headers,
-      ]),
     });
   }
 }
 
 /**
- * Parameters for the transcribe REST endpoint.
+ * List of supported STT providers and models.
  */
-export interface SpeechToTextTranscribeParams {
-  /**
-   * The format of input audio stream.
-   */
-  input_format: 'mp3' | 'wav';
-
-  /**
-   * The transcription engine to use for processing the audio stream.
-   */
-  transcription_engine: 'Azure' | 'Deepgram' | 'Google' | 'Telnyx';
-
-  /**
-   * Silence duration (in milliseconds) that triggers end-of-speech detection.
-   */
-  endpointing?: number;
-
-  /**
-   * Whether to receive interim transcription results.
-   */
-  interim_results?: boolean;
-
-  /**
-   * A key term to boost in the transcription.
-   */
-  keyterm?: string;
+export interface SpeechToTextListProvidersResponse {
+  data: Array<SpeechToTextListProvidersResponse.Data>;
 
-  /**
-   * Comma-separated list of keywords to boost in the transcription.
-   */
-  keywords?: string;
-
-  /**
-   * The language spoken in the audio stream.
-   */
-  language?: string;
-
-  /**
-   * The specific model to use within the selected transcription engine.
-   */
-  model?:
-    | 'fast'
-    | 'deepgram/nova-2'
-    | 'deepgram/nova-3'
-    | 'latest_long'
-    | 'latest_short'
-    | 'command_and_search'
-    | 'phone_call'
-    | 'video'
-    | 'default'
-    | 'medical_conversation'
-    | 'medical_dictation'
-    | 'openai/whisper-tiny'
-    | 'openai/whisper-large-v3-turbo';
-
-  /**
-   * Enable redaction of sensitive information from transcription results.
-   */
-  redact?: string;
+  meta: SpeechToTextListProvidersResponse.Meta;
 }
 
-/**
- * Parameters for establishing a speech-to-text WebSocket connection.
- */
-export interface SpeechToTextStreamParams {
+export namespace SpeechToTextListProvidersResponse {
   /**
-   * The transcription engine to use for processing the audio stream.
+   * A (provider, model) tuple along with its supported service types and languages.
    */
-  transcription_engine: 'Azure' | 'Deepgram' | 'Google' | 'Telnyx';
+  export interface Data {
+    /**
+     * Languages this (provider, model) accepts, in the provider's native code format.
+     * `auto` indicates the provider performs language detection.
+     */
+    languages: Array<string>;
 
-  /**
-   * The format of the input audio stream.
-   */
-  input_format?: 'mp3' | 'wav' | 'raw';
+    /**
+     * Provider-scoped model name.
+     */
+    model: string;
 
-  /**
-   * The language code for transcription (e.g., 'en-US', 'es-ES').
-   */
-  language?: string;
+    /**
+     * STT provider name.
+     */
+    provider: string;
 
-  /**
-   * Whether to return interim (partial) transcription results.
-   */
-  interim_results?: boolean;
+    /**
+     * Service surfaces this (provider, model) supports.
+     */
+    service_types: Array<'streaming' | 'file_transcription' | 'in_call_transcription'>;
+  }
 
-  /**
-   * The model to use for transcription (engine-specific).
-   */
-  model?: string;
+  export interface Meta {
+    /**
+     * Total number of entries returned.
+     */
+    total: number;
+  }
 }
 
 /**
- * Binary audio data sent from client to server.
- * Send raw audio bytes in mp3 or wav format.
+ * Binary audio data in mp3 or wav format.
  */
-export type SttClientEvent = ArrayBuffer | Buffer | Uint8Array;
+export type TranscribeClientEvent = Uploadable;
 
 /**
  * Union of all server-to-client WebSocket events for STT streaming.
  */
-export type SttServerEvent = SttServerEvent.TranscriptFrame | SttServerEvent.ErrorFrame;
+export type TranscribeServerEvent =
+  | TranscribeServerEvent.TranscriptFrame
+  | TranscribeServerEvent.SttErrorFrame;
 
-export namespace SttServerEvent {
+export namespace TranscribeServerEvent {
   /**
    * Server-to-client frame containing a transcription result.
    */
   export interface TranscriptFrame {
-    /**
-     * Frame type identifier.
-     */
-    type: 'transcript';
-
     /**
      * The transcribed text from the audio.
      */
     transcript: string;
 
     /**
-     * Whether this is a final transcription result.
-     * When `false`, this is an interim result that may be refined.
+     * Frame type identifier.
      */
-    is_final?: boolean;
+    type: 'transcript';
 
     /**
      * Confidence score of the transcription, ranging from 0 to 1.
      */
     confidence?: number;
+
+    /**
+     * Whether this is a final transcription result. When `false`, this is an interim
+     * result that may be refined.
+     */
+    is_final?: boolean;
   }
 
   /**
-   * Server-to-client frame indicating an error during transcription.
+   * Server-to-client frame indicating an error during transcription. The connection
+   * may be closed shortly after.
    */
-  export interface ErrorFrame {
+  export interface SttErrorFrame {
     /**
-     * Frame type identifier.
+     * Error message describing what went wrong.
      */
-    type: 'error';
+    error: string;
 
     /**
-     * Error message describing what went wrong.
+     * Frame type identifier.
      */
-    error: string;
+    type: 'error';
   }
 }
 
-/**
- * Transcript frame with transcription result.
- */
-export type TranscriptFrame = SttServerEvent.TranscriptFrame;
+export interface SpeechToTextListProvidersParams {
+  /**
+   * Filter to entries for a specific STT provider. The enum mirrors the providers
+   * advertised across the speech-to-text spec (including `google` and `telnyx`,
+   * which are accepted as WebSocket transcription engines). A provider that has no
+   * models currently registered for any service type will return an empty `data`
+   * array rather than an error.
+   */
+  provider?:
+    | 'deepgram'
+    | 'speechmatics'
+    | 'assemblyai'
+    | 'xai'
+    | 'soniox'
+    | 'azure'
+    | 'openai'
+    | 'google'
+    | 'telnyx';
 
-/**
- * Error frame indicating transcription failure.
- */
-export type SttErrorFrame = SttServerEvent.ErrorFrame;
+  /**
+   * Filter to entries that support the given service type.
+   */
+  service_type?: 'streaming' | 'file_transcription' | 'in_call_transcription';
+}
 
 export declare namespace SpeechToText {
-  export { type SpeechToTextTranscribeParams as SpeechToTextTranscribeParams };
+  export {
+    type SpeechToTextListProvidersResponse as SpeechToTextListProvidersResponse,
+    type TranscribeClientEvent as TranscribeClientEvent,
+    type TranscribeServerEvent as TranscribeServerEvent,
+    type SpeechToTextListProvidersParams as SpeechToTextListProvidersParams,
+  };
 }