SDK regeneration

.github/workflows/ci.yml

@@ -28,3 +28,30 @@ jobs:
 
       - name: Compile
         run: yarn && yarn test    
+
+  publish:
+    needs: [ compile, test ]
+    if: github.event_name == 'push' && contains(github.ref, 'refs/tags/')
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repo
+        uses: actions/checkout@v3
+      - name: Set up node
+        uses: actions/setup-node@v3
+      - name: Install dependencies
+        run: yarn install
+      - name: Build
+        run: yarn build
+
+      - name: Publish to npm
+        run: |
+          npm config set //registry.npmjs.org/:_authToken ${NPM_TOKEN}
+          if [[ ${GITHUB_REF} == *alpha* ]]; then
+            npm publish --access public --tag alpha
+          elif [[ ${GITHUB_REF} == *beta* ]]; then
+            npm publish --access public --tag beta
+          else
+            npm publish --access public
+          fi
+        env:
+          NPM_TOKEN: ${{ secrets.NPM_TOKEN }}

reference.md

@@ -44,7 +44,7 @@ await client.apiStatus.get();
 
 ## Tts
 
-<details><summary><code>client.tts.<a href="/src/api/resources/tts/client/Client.ts">bytes</a>({ ...params }) -> void</code></summary>
+<details><summary><code>client.tts.<a href="/src/api/resources/tts/client/Client.ts">bytes</a>({ ...params }) -> stream.Readable</code></summary>
 <dl>
 <dd>
 
@@ -58,17 +58,17 @@ await client.apiStatus.get();
 
 ```typescript
 await client.tts.bytes({
-    model_id: "sonic-english",
+    modelId: "sonic-english",
     transcript: "Hello, world!",
     voice: {
         mode: "id",
         id: "694f9389-aac1-45b6-b726-9d9369183238",
     },
     language: "en",
-    output_format: {
+    outputFormat: {
         container: "mp3",
-        sample_rate: 44100,
-        bit_rate: 128000,
+        sampleRate: 44100,
+        bitRate: 128000,
     },
 });
 ```
@@ -119,13 +119,18 @@ await client.tts.bytes({
 
 ```typescript
 const response = await client.tts.sse({
-    model_id: "string",
+    modelId: "string",
     transcript: "string",
     voice: {
         mode: "id",
+        id: "string",
+        experimentalControls: {
+            speed: 1.1,
+            emotion: "anger:lowest",
+        },
     },
     language: "en",
-    output_format: {
+    outputFormat: {
         container: "raw",
     },
     duration: 1.1,
@@ -200,10 +205,10 @@ This endpoint is priced at 15 characters per second of input audio.
 
 ```typescript
 await client.voiceChanger.bytes(fs.createReadStream("/path/to/your/file"), {
-    "voice[id]": "694f9389-aac1-45b6-b726-9d9369183238",
-    "output_format[container]": "mp3",
-    "output_format[sample_rate]": 44100,
-    "output_format[bit_rate]": 128000,
+    voiceId: "694f9389-aac1-45b6-b726-9d9369183238",
+    outputFormatContainer: "mp3",
+    outputFormatSampleRate: 44100,
+    outputFormatBitRate: 128000,
 });
 ```
 
@@ -261,10 +266,10 @@ await client.voiceChanger.bytes(fs.createReadStream("/path/to/your/file"), {
 
 ```typescript
 const response = await client.voiceChanger.sse(fs.createReadStream("/path/to/your/file"), {
-    "voice[id]": "694f9389-aac1-45b6-b726-9d9369183238",
-    "output_format[container]": "mp3",
-    "output_format[sample_rate]": 44100,
-    "output_format[bit_rate]": 128000,
+    voiceId: "694f9389-aac1-45b6-b726-9d9369183238",
+    outputFormatContainer: "mp3",
+    outputFormatSampleRate: 44100,
+    outputFormatBitRate: 128000,
 });
 for await (const item of response) {
     console.log(item);
@@ -378,7 +383,7 @@ await client.voices.create({
         1, 1, 1, 1, 1, 1, 1,
     ],
     language: "en",
-    base_voice_id: "string",
+    baseVoiceId: "string",
 });
 ```
 
@@ -592,7 +597,7 @@ await client.voices.localize({
         1, 1, 1, 1, 1, 1, 1,
     ],
     language: "en",
-    original_speaker_gender: "male",
+    originalSpeakerGender: "male",
     dialect: "au",
 });
 ```

src/api/resources/apiStatus/client/Client.ts

@@ -5,6 +5,7 @@
 import * as environments from "../../../../environments";
 import * as core from "../../../../core";
 import * as Cartesia from "../../../index";
+import * as serializers from "../../../../serialization/index";
 import * as errors from "../../../../errors/index";
 
 export declare namespace ApiStatus {
@@ -58,7 +59,13 @@ export class ApiStatus {
             abortSignal: requestOptions?.abortSignal,
         });
         if (_response.ok) {
-            return _response.body as Cartesia.ApiInfo;
+            return serializers.ApiInfo.parseOrThrow(_response.body, {
+                unrecognizedObjectKeys: "passthrough",
+                allowUnrecognizedUnionMembers: true,
+                allowUnrecognizedEnumValues: true,
+                skipValidation: true,
+                breadcrumbsPrefix: ["response"],
+            });
         }
 
         if (_response.error.reason === "status-code") {

src/api/resources/tts/client/Client.ts

@@ -5,10 +5,10 @@
 import * as environments from "../../../../environments";
 import * as core from "../../../../core";
 import * as Cartesia from "../../../index";
-import urlJoin from "url-join";
-import * as errors from "../../../../errors/index";
 import * as stream from "stream";
 import * as serializers from "../../../../serialization/index";
+import urlJoin from "url-join";
+import * as errors from "../../../../errors/index";
 
 export declare namespace Tts {
     interface Options {
@@ -34,60 +34,8 @@ export declare namespace Tts {
 export class Tts {
     constructor(protected readonly _options: Tts.Options = {}) {}
 
-    /**
-     * @param {Cartesia.TtsRequest} request
-     * @param {Tts.RequestOptions} requestOptions - Request-specific configuration.
-     *
-     * @example
-     *     await client.tts.bytes({
-     *         model_id: "sonic-english",
-     *         transcript: "Hello, world!",
-     *         voice: {
-     *             mode: "id",
-     *             id: "694f9389-aac1-45b6-b726-9d9369183238"
-     *         },
-     *         language: "en",
-     *         output_format: {
-     *             container: "mp3",
-     *             sample_rate: 44100,
-     *             bit_rate: 128000
-     *         }
-     *     })
-     *
-     * @example
-     *     await client.tts.bytes({
-     *         model_id: "sonic-english",
-     *         transcript: "Hello, world!",
-     *         voice: {
-     *             mode: "id",
-     *             id: "694f9389-aac1-45b6-b726-9d9369183238"
-     *         },
-     *         language: "en",
-     *         output_format: {
-     *             container: "wav",
-     *             sample_rate: 44100,
-     *             encoding: "pcm_f32le"
-     *         }
-     *     })
-     *
-     * @example
-     *     await client.tts.bytes({
-     *         model_id: "sonic-english",
-     *         transcript: "Hello, world!",
-     *         voice: {
-     *             mode: "id",
-     *             id: "694f9389-aac1-45b6-b726-9d9369183238"
-     *         },
-     *         language: "en",
-     *         output_format: {
-     *             container: "raw",
-     *             sample_rate: 44100,
-     *             encoding: "pcm_f32le"
-     *         }
-     *     })
-     */
-    public async bytes(request: Cartesia.TtsRequest, requestOptions?: Tts.RequestOptions): Promise<void> {
-        const _response = await (this._options.fetcher ?? core.fetcher)({
+    public async bytes(request: Cartesia.TtsRequest, requestOptions?: Tts.RequestOptions): Promise<stream.Readable> {
+        const _response = await (this._options.fetcher ?? core.fetcher)<stream.Readable>({
             url: urlJoin(
                 (await core.Supplier.get(this._options.environment)) ?? environments.CartesiaEnvironment.Production,
                 "/tts/bytes"
@@ -105,13 +53,14 @@ export class Tts {
             },
             contentType: "application/json",
             requestType: "json",
-            body: request,
+            body: serializers.TtsRequest.jsonOrThrow(request, { unrecognizedObjectKeys: "strip" }),
+            responseType: "streaming",
             timeoutMs: requestOptions?.timeoutInSeconds != null ? requestOptions.timeoutInSeconds * 1000 : 60000,
             maxRetries: requestOptions?.maxRetries,
             abortSignal: requestOptions?.abortSignal,
         });
         if (_response.ok) {
-            return;
+            return _response.body;
         }
 
         if (_response.error.reason === "status-code") {
@@ -158,7 +107,7 @@ export class Tts {
             },
             contentType: "application/json",
             requestType: "json",
-            body: request,
+            body: serializers.TtsRequest.jsonOrThrow(request, { unrecognizedObjectKeys: "strip" }),
             responseType: "sse",
             timeoutMs: requestOptions?.timeoutInSeconds != null ? requestOptions.timeoutInSeconds * 1000 : 60000,
             maxRetries: requestOptions?.maxRetries,
@@ -178,8 +127,8 @@ export class Tts {
                 },
                 signal: requestOptions?.abortSignal,
                 eventShape: {
-                    type: "json",
-                    messageTerminator: "\n",
+                    type: "sse",
+                    streamTerminator: "[DONE]",
                 },
             });
         }

src/api/resources/tts/types/CancelContextRequest.ts

@@ -6,7 +6,7 @@ import * as Cartesia from "../../../index";
 
 export interface CancelContextRequest {
     /** The ID of the context to cancel. */
-    context_id: Cartesia.ContextId;
+    contextId: Cartesia.ContextId;
     /** Whether to cancel the context, so that no more messages are generated for that context. */
     cancel: true;
 }

src/api/resources/tts/types/GenerationRequest.ts

@@ -6,22 +6,22 @@ import * as Cartesia from "../../../index";
 
 export interface GenerationRequest {
     /** The ID of the model to use for the generation. See [Models](/build-with-sonic/models) for available models. */
-    model_id: string;
+    modelId: string;
     transcript: string;
     voice: Cartesia.TtsRequestVoiceSpecifier;
     language?: Cartesia.SupportedLanguage;
-    output_format: Cartesia.WebSocketRawOutputFormat;
+    outputFormat: Cartesia.WebSocketRawOutputFormat;
     /**
      * The maximum duration of the audio in seconds. You do not usually need to specify this.
      * If the duration is not appropriate for the length of the transcript, the output audio may be truncated.
      */
     duration?: number;
-    context_id: Cartesia.ContextId;
+    contextId: Cartesia.ContextId;
     /**
      * Whether this input may be followed by more inputs.
      * If not specified, this defaults to `false`.
      */
     continue?: boolean;
     /** Whether to return word-level timestamps. */
-    add_timestamps?: boolean;
+    addTimestamps?: boolean;
 }

src/api/resources/tts/types/Mp3OutputFormat.ts

@@ -3,6 +3,6 @@
  */
 
 export interface Mp3OutputFormat {
-    sample_rate: number;
-    bit_rate: number;
+    sampleRate: number;
+    bitRate: number;
 }

src/api/resources/tts/types/RawOutputFormat.ts

@@ -6,5 +6,5 @@ import * as Cartesia from "../../../index";
 
 export interface RawOutputFormat {
     encoding: Cartesia.RawEncoding;
-    sample_rate: number;
+    sampleRate: number;
 }

src/api/resources/tts/types/TtsRequest.ts

@@ -6,11 +6,11 @@ import * as Cartesia from "../../../index";
 
 export interface TtsRequest {
     /** The ID of the model to use for the generation. See [Models](/build-with-sonic/models) for available models. */
-    model_id: string;
+    modelId: string;
     transcript: string;
     voice: Cartesia.TtsRequestVoiceSpecifier;
     language?: Cartesia.SupportedLanguage;
-    output_format: Cartesia.OutputFormat;
+    outputFormat: Cartesia.OutputFormat;
     /**
      * The maximum duration of the audio in seconds. You do not usually need to specify this.
      * If the duration is not appropriate for the length of the transcript, the output audio may be truncated.

src/api/resources/tts/types/TtsRequestEmbeddingSpecifier.ts

@@ -5,6 +5,7 @@
 import * as Cartesia from "../../../index";
 
 export interface TtsRequestEmbeddingSpecifier {
+    mode: "embedding";
     embedding: Cartesia.Embedding;
-    __experimental_controls?: Cartesia.Controls;
+    experimentalControls?: Cartesia.Controls;
 }

src/api/resources/tts/types/TtsRequestIdSpecifier.ts

@@ -5,6 +5,7 @@
 import * as Cartesia from "../../../index";
 
 export interface TtsRequestIdSpecifier {
+    mode: "id";
     id: Cartesia.VoiceId;
-    __experimental_controls?: Cartesia.Controls;
+    experimentalControls?: Cartesia.Controls;
 }

src/api/resources/tts/types/TtsRequestVoiceSpecifier.ts

@@ -4,16 +4,4 @@
 
 import * as Cartesia from "../../../index";
 
-export type TtsRequestVoiceSpecifier =
-    | Cartesia.TtsRequestVoiceSpecifier.Id
-    | Cartesia.TtsRequestVoiceSpecifier.Embedding;
-
-export declare namespace TtsRequestVoiceSpecifier {
-    interface Id extends Cartesia.TtsRequestIdSpecifier {
-        mode: "id";
-    }
-
-    interface Embedding extends Cartesia.TtsRequestEmbeddingSpecifier {
-        mode: "embedding";
-    }
-}
+export type TtsRequestVoiceSpecifier = Cartesia.TtsRequestIdSpecifier | Cartesia.TtsRequestEmbeddingSpecifier;

src/api/resources/tts/types/WebSocketBaseResponse.ts

@@ -5,7 +5,7 @@
 import * as Cartesia from "../../../index";
 
 export interface WebSocketBaseResponse {
-    context_id: Cartesia.ContextId;
-    status_code: number;
+    contextId?: Cartesia.ContextId;
+    statusCode: number;
     done: boolean;
 }

src/api/resources/tts/types/WebSocketChunkResponse.ts

@@ -6,5 +6,5 @@ import * as Cartesia from "../../../index";
 
 export interface WebSocketChunkResponse extends Cartesia.WebSocketBaseResponse {
     data: string;
-    step_time: number;
+    stepTime: number;
 }

src/api/resources/tts/types/WebSocketRawOutputFormat.ts

@@ -7,5 +7,5 @@ import * as Cartesia from "../../../index";
 export interface WebSocketRawOutputFormat {
     container: "raw";
     encoding: Cartesia.RawEncoding;
-    sample_rate: number;
+    sampleRate: number;
 }