@meetkai/mka1 0.55.18 → 0.55.20

package/src/models/operations/listmodels.ts

@@ -69,6 +69,17 @@ export const ListModelsWeights = {
 } as const;
 export type ListModelsWeights = ClosedEnum<typeof ListModelsWeights>;
 
+export type TtsVoices = {
+  id: string;
+  name: string;
+  language: string;
+};
+
+export type ListModelsTts = {
+  defaultVoice: { [k: string]: string };
+  voices: Array<TtsVoices>;
+};
+
 /**
  * `active`, `deprecated` (still served, retirement announced or pending) or `retired` (the retirement date has passed). Informational in this version: the gateway does not yet refuse a retired model.
  */
@@ -196,6 +207,10 @@ export type ListModelsData = {
    * Languages the model is intended for, as ISO 639-1 codes, mirroring `modelCard.languages`. Null when unrecorded.
    */
   languages: Array<string> | null;
+  /**
+   * Available public TTS voices and default voice IDs by language. Provider voice identifiers are not exposed.
+   */
+  tts: ListModelsTts | null;
   /**
    * Known limitations in one short paragraph, mirroring `modelCard.limitations`. Null when unrecorded.
    */
@@ -418,6 +433,93 @@ export const ListModelsWeights$outboundSchema: z.ZodNativeEnum<
   typeof ListModelsWeights
 > = ListModelsWeights$inboundSchema;
 
+/** @internal */
+export const TtsVoices$inboundSchema: z.ZodType<
+  TtsVoices,
+  z.ZodTypeDef,
+  unknown
+> = z.object({
+  id: z.string(),
+  name: z.string(),
+  language: z.string(),
+});
+/** @internal */
+export type TtsVoices$Outbound = {
+  id: string;
+  name: string;
+  language: string;
+};
+
+/** @internal */
+export const TtsVoices$outboundSchema: z.ZodType<
+  TtsVoices$Outbound,
+  z.ZodTypeDef,
+  TtsVoices
+> = z.object({
+  id: z.string(),
+  name: z.string(),
+  language: z.string(),
+});
+
+export function ttsVoicesToJSON(ttsVoices: TtsVoices): string {
+  return JSON.stringify(TtsVoices$outboundSchema.parse(ttsVoices));
+}
+export function ttsVoicesFromJSON(
+  jsonString: string,
+): SafeParseResult<TtsVoices, SDKValidationError> {
+  return safeParse(
+    jsonString,
+    (x) => TtsVoices$inboundSchema.parse(JSON.parse(x)),
+    `Failed to parse 'TtsVoices' from JSON`,
+  );
+}
+
+/** @internal */
+export const ListModelsTts$inboundSchema: z.ZodType<
+  ListModelsTts,
+  z.ZodTypeDef,
+  unknown
+> = z.object({
+  default_voice: z.record(z.string()),
+  voices: z.array(z.lazy(() => TtsVoices$inboundSchema)),
+}).transform((v) => {
+  return remap$(v, {
+    "default_voice": "defaultVoice",
+  });
+});
+/** @internal */
+export type ListModelsTts$Outbound = {
+  default_voice: { [k: string]: string };
+  voices: Array<TtsVoices$Outbound>;
+};
+
+/** @internal */
+export const ListModelsTts$outboundSchema: z.ZodType<
+  ListModelsTts$Outbound,
+  z.ZodTypeDef,
+  ListModelsTts
+> = z.object({
+  defaultVoice: z.record(z.string()),
+  voices: z.array(z.lazy(() => TtsVoices$outboundSchema)),
+}).transform((v) => {
+  return remap$(v, {
+    defaultVoice: "default_voice",
+  });
+});
+
+export function listModelsTtsToJSON(listModelsTts: ListModelsTts): string {
+  return JSON.stringify(ListModelsTts$outboundSchema.parse(listModelsTts));
+}
+export function listModelsTtsFromJSON(
+  jsonString: string,
+): SafeParseResult<ListModelsTts, SDKValidationError> {
+  return safeParse(
+    jsonString,
+    (x) => ListModelsTts$inboundSchema.parse(JSON.parse(x)),
+    `Failed to parse 'ListModelsTts' from JSON`,
+  );
+}
+
 /** @internal */
 export const ListModelsStatus$inboundSchema: z.ZodNativeEnum<
   typeof ListModelsStatus
@@ -584,6 +686,7 @@ export const ListModelsData$inboundSchema: z.ZodType<
   weights: z.nullable(ListModelsWeights$inboundSchema),
   license: z.nullable(z.string()),
   languages: z.nullable(z.array(z.string())),
+  tts: z.nullable(z.lazy(() => ListModelsTts$inboundSchema)),
   limitations: z.nullable(z.string()),
   lifecycle: z.lazy(() => ListModelsLifecycle$inboundSchema),
   capabilities: z.lazy(() => ListModelsCapabilities$inboundSchema),
@@ -626,6 +729,7 @@ export type ListModelsData$Outbound = {
   weights: string | null;
   license: string | null;
   languages: Array<string> | null;
+  tts: ListModelsTts$Outbound | null;
   limitations: string | null;
   lifecycle: ListModelsLifecycle$Outbound;
   capabilities: ListModelsCapabilities$Outbound;
@@ -663,6 +767,7 @@ export const ListModelsData$outboundSchema: z.ZodType<
   weights: z.nullable(ListModelsWeights$outboundSchema),
   license: z.nullable(z.string()),
   languages: z.nullable(z.array(z.string())),
+  tts: z.nullable(z.lazy(() => ListModelsTts$outboundSchema)),
   limitations: z.nullable(z.string()),
   lifecycle: z.lazy(() => ListModelsLifecycle$outboundSchema),
   capabilities: z.lazy(() => ListModelsCapabilities$outboundSchema),

package/src/react-query/llmSpeechGetTranscriptionHistoryContent.core.ts

@@ -8,20 +8,24 @@ import {
   QueryKey,
 } from "@tanstack/react-query";
 import { SDKCore } from "../core.js";
-import { llmSpeechGetTranscriptionHistoryContent } from "../funcs/llmSpeechGetTranscriptionHistoryContent.js";
+import {
+  GetTranscriptionHistoryContentAcceptEnum,
+  llmSpeechGetTranscriptionHistoryContent,
+} from "../funcs/llmSpeechGetTranscriptionHistoryContent.js";
 import { combineSignals } from "../lib/primitives.js";
 import { RequestOptions } from "../lib/sdks.js";
 import * as operations from "../models/operations/index.js";
 import { unwrapAsync } from "../types/fp.js";
-export type LlmSpeechGetTranscriptionHistoryContentQueryData = ReadableStream<
-  Uint8Array
->;
+export type LlmSpeechGetTranscriptionHistoryContentQueryData =
+  operations.GetSpeechToTextHistoryContentResponse;
 
 export function prefetchLlmSpeechGetTranscriptionHistoryContent(
   queryClient: QueryClient,
   client$: SDKCore,
   request: operations.GetSpeechToTextHistoryContentRequest,
-  options?: RequestOptions,
+  options?: RequestOptions & {
+    acceptHeaderOverride?: GetTranscriptionHistoryContentAcceptEnum;
+  },
 ): Promise<void> {
   return queryClient.prefetchQuery({
     ...buildLlmSpeechGetTranscriptionHistoryContentQuery(
@@ -35,7 +39,9 @@ export function prefetchLlmSpeechGetTranscriptionHistoryContent(
 export function buildLlmSpeechGetTranscriptionHistoryContentQuery(
   client$: SDKCore,
   request: operations.GetSpeechToTextHistoryContentRequest,
-  options?: RequestOptions,
+  options?: RequestOptions & {
+    acceptHeaderOverride?: GetTranscriptionHistoryContentAcceptEnum;
+  },
 ): {
   queryKey: QueryKey;
   queryFn: (

package/src/react-query/llmSpeechGetTranscriptionHistoryContent.ts

@@ -50,6 +50,8 @@ export type LlmSpeechGetTranscriptionHistoryContentQueryError =
   | UnexpectedClientError
   | SDKValidationError;
 
+export { GetTranscriptionHistoryContentAcceptEnum } from "../funcs/llmSpeechGetTranscriptionHistoryContent.js";
+
 /**
  * Retrieve speech-to-text audio
  *

package/src/react-query/llmSpeechSpeak.ts

@@ -52,6 +52,7 @@ export type LlmSpeechSpeakMutationError =
  * **Request Body:**
  * - `text`: Input text to convert to speech - required
  * - `language`: Language code (default: "auto") - "auto" for automatic detection, or one of the supported ISO 639-1 codes: en, zh, hi, es, ar, bn, pt, ru, ja, pa, de, ko, fr, tr, it, th, pl, nl, id, vi, ur, uk
+ * - `voice`: Optional public voice ID from model discovery; otherwise the resolved language selects its configured default. The example assumes an operator-configured `meetkai:speech` model exposing `sophia`; use IDs advertised by your deployment
  * - `model`: TTS model identifier (default: "auto") - "auto" selects the default provider; a registry-resolvable id dispatches to the matching provider
  *
  * **Response:** Returns audio file in WAV format with X-Language-Code header. When `store` is `true`, the generation is persisted before the response returns and the new history record id is included in the X-Tts-Id header.

package/src/react-query/llmSpeechSpeakStreaming.ts

@@ -67,6 +67,7 @@ export type LlmSpeechSpeakStreamingMutationError =
  * - `text`: Input text to convert to speech - required
  * - `language`: Language code (default: "auto") - "auto" for automatic detection, or one of the supported ISO 639-1 codes: en, zh, hi, es, ar, bn, pt, ru, ja, pa, de, ko, fr, tr, it, th, pl, nl, id, vi, ur, uk
  * - `format`: Audio format (default: "mp3") - "mp3" for compressed MPEG audio (96 kbps) or "pcm" for uncompressed WAV
+ * - `voice`: Optional public voice ID from model discovery; otherwise the resolved language selects its configured default. The example assumes an operator-configured `meetkai:speech` model exposing `sophia`; use IDs advertised by your deployment
  * - `model`: TTS model identifier (default: "auto") - "auto" selects the default provider; a registry-resolvable id dispatches to the matching provider (note: not all providers support `format: "mp3"`)
  *
  * **Response:**

package/src/sdk/speech.ts

@@ -5,7 +5,10 @@
 import { llmSpeechDeleteTranscriptionHistory } from "../funcs/llmSpeechDeleteTranscriptionHistory.js";
 import { llmSpeechDeleteTtsHistory } from "../funcs/llmSpeechDeleteTtsHistory.js";
 import { llmSpeechGetTranscriptionHistory } from "../funcs/llmSpeechGetTranscriptionHistory.js";
-import { llmSpeechGetTranscriptionHistoryContent } from "../funcs/llmSpeechGetTranscriptionHistoryContent.js";
+import {
+  GetTranscriptionHistoryContentAcceptEnum,
+  llmSpeechGetTranscriptionHistoryContent,
+} from "../funcs/llmSpeechGetTranscriptionHistoryContent.js";
 import { llmSpeechGetTtsHistory } from "../funcs/llmSpeechGetTtsHistory.js";
 import {
   GetTtsHistoryContentAcceptEnum,
@@ -29,6 +32,8 @@ export { SpeakStreamingAcceptEnum } from "../funcs/llmSpeechSpeakStreaming.js";
 
 export { GetTtsHistoryContentAcceptEnum } from "../funcs/llmSpeechGetTtsHistoryContent.js";
 
+export { GetTranscriptionHistoryContentAcceptEnum } from "../funcs/llmSpeechGetTranscriptionHistoryContent.js";
+
 export class Speech extends ClientSDK {
   /**
    * Speech to text transcription
@@ -87,6 +92,7 @@ export class Speech extends ClientSDK {
    * **Request Body:**
    * - `text`: Input text to convert to speech - required
    * - `language`: Language code (default: "auto") - "auto" for automatic detection, or one of the supported ISO 639-1 codes: en, zh, hi, es, ar, bn, pt, ru, ja, pa, de, ko, fr, tr, it, th, pl, nl, id, vi, ur, uk
+   * - `voice`: Optional public voice ID from model discovery; otherwise the resolved language selects its configured default. The example assumes an operator-configured `meetkai:speech` model exposing `sophia`; use IDs advertised by your deployment
    * - `model`: TTS model identifier (default: "auto") - "auto" selects the default provider; a registry-resolvable id dispatches to the matching provider
    *
    * **Response:** Returns audio file in WAV format with X-Language-Code header. When `store` is `true`, the generation is persisted before the response returns and the new history record id is included in the X-Tts-Id header.
@@ -118,6 +124,7 @@ export class Speech extends ClientSDK {
    * - `text`: Input text to convert to speech - required
    * - `language`: Language code (default: "auto") - "auto" for automatic detection, or one of the supported ISO 639-1 codes: en, zh, hi, es, ar, bn, pt, ru, ja, pa, de, ko, fr, tr, it, th, pl, nl, id, vi, ur, uk
    * - `format`: Audio format (default: "mp3") - "mp3" for compressed MPEG audio (96 kbps) or "pcm" for uncompressed WAV
+   * - `voice`: Optional public voice ID from model discovery; otherwise the resolved language selects its configured default. The example assumes an operator-configured `meetkai:speech` model exposing `sophia`; use IDs advertised by your deployment
    * - `model`: TTS model identifier (default: "auto") - "auto" selects the default provider; a registry-resolvable id dispatches to the matching provider (note: not all providers support `format: "mp3"`)
    *
    * **Response:**
@@ -290,8 +297,10 @@ export class Speech extends ClientSDK {
    */
   async getTranscriptionHistoryContent(
     request: operations.GetSpeechToTextHistoryContentRequest,
-    options?: RequestOptions,
-  ): Promise<ReadableStream<Uint8Array>> {
+    options?: RequestOptions & {
+      acceptHeaderOverride?: GetTranscriptionHistoryContentAcceptEnum;
+    },
+  ): Promise<operations.GetSpeechToTextHistoryContentResponse> {
     return unwrapAsync(llmSpeechGetTranscriptionHistoryContent(
       this,
       request,