feat(api): add audio create method (#127)

Author: stainless-app[bot]

.stats.yml

@@ -1,2 +1,2 @@
-configured_endpoints: 16
+configured_endpoints: 17
 openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/togetherai%2FTogetherAI-87bcc1e47f49a52fc31601e87d387cf7ef17f50fba0a06aa105a7fcc65644de4.yml

README.md

@@ -4,7 +4,7 @@
 
 This library provides convenient access to the Together REST API from server-side TypeScript or JavaScript.
 
-The REST API documentation can be found on [docs.together.ai](https://docs.together.ai). The full API of this library can be found in [api.md](api.md).
+The REST API documentation can be found on [docs.together.ai](https://docs.together.ai/). The full API of this library can be found in [api.md](api.md).
 
 It is generated with [Stainless](https://www.stainlessapi.com/).
 

api.md

@@ -99,6 +99,16 @@ Methods:
 
 - <code title="post /images/generations">client.images.<a href="./src/resources/images.ts">create</a>({ ...params }) -> ImageFile</code>
 
+# Audio
+
+Types:
+
+- <code><a href="./src/resources/audio.ts">AudioFile</a></code>
+
+Methods:
+
+- <code title="post /audio/speech">client.audio.<a href="./src/resources/audio.ts">create</a>({ ...params }) -> Response</code>
+
 # Models
 
 Types:

src/index.ts

@@ -7,6 +7,7 @@ import * as Uploads from './uploads';
 import * as API from './resources/index';
 import * as TopLevelAPI from './resources/top-level';
 import { RerankParams, RerankResponse } from './resources/top-level';
+import { Audio, AudioCreateParams, AudioFile } from './resources/audio';
 import {
   Completion,
   CompletionCreateParams,
@@ -157,6 +158,7 @@ export class Together extends Core.APIClient {
   files: API.Files = new API.Files(this);
   fineTune: API.FineTuneResource = new API.FineTuneResource(this);
   images: API.Images = new API.Images(this);
+  audio: API.Audio = new API.Audio(this);
   models: API.Models = new API.Models(this);
 
   /**
@@ -211,6 +213,7 @@ Together.Embeddings = Embeddings;
 Together.Files = Files;
 Together.FineTuneResource = FineTuneResource;
 Together.Images = Images;
+Together.Audio = Audio;
 Together.Models = Models;
 export declare namespace Together {
   export type RequestOptions = Core.RequestOptions;
@@ -256,6 +259,8 @@ export declare namespace Together {
 
   export { Images as Images, type ImageFile as ImageFile, type ImageCreateParams as ImageCreateParams };
 
+  export { Audio as Audio, type AudioFile as AudioFile, type AudioCreateParams as AudioCreateParams };
+
   export { Models as Models, type ModelListResponse as ModelListResponse };
 }
 

src/resources/audio.ts

@@ -0,0 +1,110 @@
+// File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+import { APIResource } from '../resource';
+import * as Core from '../core';
+import { type Response } from '../_shims/index';
+
+export class Audio extends APIResource {
+  /**
+   * Generate audio from input text
+   */
+  create(body: AudioCreateParams, options?: Core.RequestOptions): Core.APIPromise<Response> {
+    return this._client.post('/audio/speech', {
+      body,
+      ...options,
+      headers: { Accept: 'application/octet-stream', ...options?.headers },
+      __binaryResponse: true,
+    });
+  }
+}
+
+export type AudioFile = AudioFile.AudioSpeechStreamEvent | AudioFile.StreamSentinel;
+
+export namespace AudioFile {
+  export interface AudioSpeechStreamEvent {
+    data: AudioSpeechStreamEvent.Data;
+  }
+
+  export namespace AudioSpeechStreamEvent {
+    export interface Data {
+      /**
+       * base64 encoded audio stream
+       */
+      b64: string;
+
+      model: string;
+
+      object: 'audio.tts.chunk';
+    }
+  }
+
+  export interface StreamSentinel {
+    data: '[DONE]';
+  }
+}
+
+export interface AudioCreateParams {
+  /**
+   * Input text to generate the audio for
+   */
+  input: string;
+
+  /**
+   * The name of the model to query.
+   *
+   * [See all of Together AI's chat models](https://docs.together.ai/docs/serverless-models#audio-models)
+   */
+  model: 'cartesia/sonic' | (string & {});
+
+  /**
+   * The voice to use for generating the audio.
+   * [View all supported voices here](https://docs.together.ai/docs/text-to-speech#voices-available).
+   */
+  voice: 'laidback woman' | 'polite man' | 'storyteller lady' | 'friendly sidekick' | (string & {});
+
+  /**
+   * Language of input text
+   */
+  language?:
+    | 'en'
+    | 'de'
+    | 'fr'
+    | 'es'
+    | 'hi'
+    | 'it'
+    | 'ja'
+    | 'ko'
+    | 'nl'
+    | 'pl'
+    | 'pt'
+    | 'ru'
+    | 'sv'
+    | 'tr'
+    | 'zh';
+
+  /**
+   * Audio encoding of response
+   */
+  response_encoding?: 'pcm_f32le' | 'pcm_s16le' | 'pcm_mulaw' | 'pcm_alaw';
+
+  /**
+   * The format of audio output
+   */
+  response_format?: 'mp3' | 'wav' | 'raw';
+
+  /**
+   * Sampling rate to use for the output audio
+   */
+  sample_rate?: number;
+
+  /**
+   * If true, output is streamed for several characters at a time instead of waiting
+   * for the full response. The stream terminates with `data: [DONE]`. If false,
+   * return the encoded audio as octet stream
+   */
+  stream?: boolean;
+}
+
+export declare namespace Audio {
+  export { type AudioFile as AudioFile, type AudioCreateParams as AudioCreateParams };
+}

src/resources/index.ts

@@ -1,5 +1,6 @@
 // File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
+export { Audio, type AudioFile, type AudioCreateParams } from './audio';
 export { Chat } from './chat/chat';
 export {
   Completions,

tests/api-resources/audio.test.ts

@@ -0,0 +1,23 @@
+// File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+import Together from 'together-ai';
+
+const client = new Together({
+  apiKey: 'My API Key',
+  baseURL: process.env['TEST_API_BASE_URL'] ?? 'http://127.0.0.1:4010',
+});
+
+describe('resource audio', () => {
+  test('create: required and optional params', async () => {
+    const response = await client.audio.create({
+      input: 'input',
+      model: 'cartesia/sonic',
+      voice: 'laidback woman',
+      language: 'en',
+      response_encoding: 'pcm_f32le',
+      response_format: 'mp3',
+      sample_rate: 0,
+      stream: true,
+    });
+  });
+});