@cartesia/cartesia-js 2.2.5 → 2.2.7

package/api/resources/apiStatus/client/Client.js

@@ -66,7 +66,7 @@ class ApiStatus {
             const _response = yield ((_a = this._options.fetcher) !== null && _a !== void 0 ? _a : core.fetcher)({
                 url: (_c = (_b = (yield core.Supplier.get(this._options.baseUrl))) !== null && _b !== void 0 ? _b : (yield core.Supplier.get(this._options.environment))) !== null && _c !== void 0 ? _c : environments.CartesiaEnvironment.Production,
                 method: "GET",
-                headers: Object.assign(Object.assign({ "Cartesia-Version": (_f = (_d = requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.cartesiaVersion) !== null && _d !== void 0 ? _d : (_e = this._options) === null || _e === void 0 ? void 0 : _e.cartesiaVersion) !== null && _f !== void 0 ? _f : "2024-06-10", "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "@cartesia/cartesia-js", "X-Fern-SDK-Version": "2.2.5", "User-Agent": "@cartesia/cartesia-js/2.2.5", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())), requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.headers),
+                headers: Object.assign(Object.assign({ "Cartesia-Version": (_f = (_d = requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.cartesiaVersion) !== null && _d !== void 0 ? _d : (_e = this._options) === null || _e === void 0 ? void 0 : _e.cartesiaVersion) !== null && _f !== void 0 ? _f : "2024-06-10", "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "@cartesia/cartesia-js", "X-Fern-SDK-Version": "2.2.7", "User-Agent": "@cartesia/cartesia-js/2.2.7", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())), requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.headers),
                 contentType: "application/json",
                 requestType: "json",
                 timeoutMs: (requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) != null ? requestOptions.timeoutInSeconds * 1000 : 60000,

package/api/resources/auth/client/Client.d.ts

@@ -39,10 +39,27 @@ export declare class Auth {
      * @example
      *     await client.auth.accessToken({
      *         grants: {
+     *             tts: true,
+     *             stt: true
+     *         },
+     *         expiresIn: 60
+     *     })
+     *
+     * @example
+     *     await client.auth.accessToken({
+     *         grants: {
      *             tts: true
      *         },
      *         expiresIn: 60
      *     })
+     *
+     * @example
+     *     await client.auth.accessToken({
+     *         grants: {
+     *             stt: true
+     *         },
+     *         expiresIn: 60
+     *     })
      */
     accessToken(request: Cartesia.TokenRequest, requestOptions?: Auth.RequestOptions): Promise<Cartesia.TokenResponse>;
     protected _getCustomAuthorizationHeaders(): Promise<{

package/api/resources/auth/client/Client.js

@@ -67,10 +67,27 @@ class Auth {
      * @example
      *     await client.auth.accessToken({
      *         grants: {
+     *             tts: true,
+     *             stt: true
+     *         },
+     *         expiresIn: 60
+     *     })
+     *
+     * @example
+     *     await client.auth.accessToken({
+     *         grants: {
      *             tts: true
      *         },
      *         expiresIn: 60
      *     })
+     *
+     * @example
+     *     await client.auth.accessToken({
+     *         grants: {
+     *             stt: true
+     *         },
+     *         expiresIn: 60
+     *     })
      */
     accessToken(request, requestOptions) {
         return __awaiter(this, void 0, void 0, function* () {
@@ -78,7 +95,7 @@ class Auth {
             const _response = yield ((_a = this._options.fetcher) !== null && _a !== void 0 ? _a : core.fetcher)({
                 url: (0, url_join_1.default)((_c = (_b = (yield core.Supplier.get(this._options.baseUrl))) !== null && _b !== void 0 ? _b : (yield core.Supplier.get(this._options.environment))) !== null && _c !== void 0 ? _c : environments.CartesiaEnvironment.Production, "access-token"),
                 method: "POST",
-                headers: Object.assign(Object.assign({ "Cartesia-Version": (_f = (_d = requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.cartesiaVersion) !== null && _d !== void 0 ? _d : (_e = this._options) === null || _e === void 0 ? void 0 : _e.cartesiaVersion) !== null && _f !== void 0 ? _f : "2024-06-10", "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "@cartesia/cartesia-js", "X-Fern-SDK-Version": "2.2.5", "User-Agent": "@cartesia/cartesia-js/2.2.5", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())), requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.headers),
+                headers: Object.assign(Object.assign({ "Cartesia-Version": (_f = (_d = requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.cartesiaVersion) !== null && _d !== void 0 ? _d : (_e = this._options) === null || _e === void 0 ? void 0 : _e.cartesiaVersion) !== null && _f !== void 0 ? _f : "2024-06-10", "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "@cartesia/cartesia-js", "X-Fern-SDK-Version": "2.2.7", "User-Agent": "@cartesia/cartesia-js/2.2.7", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())), requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.headers),
                 contentType: "application/json",
                 requestType: "json",
                 body: serializers.TokenRequest.jsonOrThrow(request, { unrecognizedObjectKeys: "strip" }),

package/api/resources/auth/types/TokenGrant.d.ts

@@ -3,5 +3,7 @@
  */
 export interface TokenGrant {
     /** The `tts` grant allows the token to be used to access any TTS endpoint. */
-    tts: boolean;
+    tts?: boolean;
+    /** The `stt` grant allows the token to be used to access any STT endpoint. */
+    stt?: boolean;
 }

package/api/resources/auth/types/TokenRequest.d.ts

@@ -3,8 +3,8 @@
  */
 import * as Cartesia from "../../../index";
 export interface TokenRequest {
-    /** The permissions to be granted via the token. */
-    grants: Cartesia.TokenGrant;
+    /** The permissions to be granted via the token. Both TTS and STT grants are optional - specify only the capabilities you need. */
+    grants?: Cartesia.TokenGrant;
     /** The number of seconds the token will be valid for since the time of generation. The maximum is 1 hour (3600 seconds). */
     expiresIn?: number;
 }

package/api/resources/index.d.ts

@@ -14,5 +14,6 @@ export * as voices from "./voices";
 export * from "./voices/types";
 export * as infill from "./infill";
 export * from "./infill/client/requests";
+export * from "./stt/client/requests";
 export * from "./voiceChanger/client/requests";
 export * from "./voices/client/requests";

package/api/resources/index.js

@@ -53,5 +53,6 @@ exports.voices = __importStar(require("./voices"));
 __exportStar(require("./voices/types"), exports);
 exports.infill = __importStar(require("./infill"));
 __exportStar(require("./infill/client/requests"), exports);
+__exportStar(require("./stt/client/requests"), exports);
 __exportStar(require("./voiceChanger/client/requests"), exports);
 __exportStar(require("./voices/client/requests"), exports);

package/api/resources/infill/client/Client.js

@@ -113,7 +113,7 @@ class Infill {
             const _response = yield ((_a = this._options.fetcher) !== null && _a !== void 0 ? _a : core.fetcher)({
                 url: (0, url_join_1.default)((_c = (_b = (yield core.Supplier.get(this._options.baseUrl))) !== null && _b !== void 0 ? _b : (yield core.Supplier.get(this._options.environment))) !== null && _c !== void 0 ? _c : environments.CartesiaEnvironment.Production, "/infill/bytes"),
                 method: "POST",
-                headers: Object.assign(Object.assign(Object.assign({ "Cartesia-Version": (_f = (_d = requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.cartesiaVersion) !== null && _d !== void 0 ? _d : (_e = this._options) === null || _e === void 0 ? void 0 : _e.cartesiaVersion) !== null && _f !== void 0 ? _f : "2024-06-10", "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "@cartesia/cartesia-js", "X-Fern-SDK-Version": "2.2.5", "User-Agent": "@cartesia/cartesia-js/2.2.5", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())), _maybeEncodedRequest.headers), requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.headers),
+                headers: Object.assign(Object.assign(Object.assign({ "Cartesia-Version": (_f = (_d = requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.cartesiaVersion) !== null && _d !== void 0 ? _d : (_e = this._options) === null || _e === void 0 ? void 0 : _e.cartesiaVersion) !== null && _f !== void 0 ? _f : "2024-06-10", "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "@cartesia/cartesia-js", "X-Fern-SDK-Version": "2.2.7", "User-Agent": "@cartesia/cartesia-js/2.2.7", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())), _maybeEncodedRequest.headers), requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.headers),
                 requestType: "file",
                 duplex: _maybeEncodedRequest.duplex,
                 body: _maybeEncodedRequest.body,

package/api/resources/infill/client/requests/InfillBytesRequest.d.ts

@@ -40,23 +40,16 @@ export interface InfillBytesRequest {
     voiceId: string;
     /** The format of the output audio */
     outputFormatContainer: Cartesia.OutputFormatContainer;
-    /** The sample rate of the output audio */
+    /** The sample rate of the output audio in Hz. Supported sample rates are 8000, 16000, 22050, 24000, 44100, 48000. */
     outputFormatSampleRate: number;
-    /**
-     * Required for `raw` and `wav` containers.
-     *
-     */
+    /** Required for `raw` and `wav` containers. */
     outputFormatEncoding?: Cartesia.RawEncoding;
-    /**
-     * Required for `mp3` containers.
-     *
-     */
+    /** Required for `mp3` containers. */
     outputFormatBitRate?: number;
     /**
      * Either a number between -1.0 and 1.0 or a natural language description of speed.
      *
      * If you specify a number, 0.0 is the default speed, -1.0 is the slowest speed, and 1.0 is the fastest speed.
-     *
      */
     voiceExperimentalControlsSpeed?: Cartesia.Speed;
     /**
@@ -65,7 +58,6 @@ export interface InfillBytesRequest {
      * Supported emotions are: anger, positivity, surprise, sadness, and curiosity.
      *
      * Supported levels are: lowest, low, (omit), high, highest.
-     *
      */
     voiceExperimentalControlsEmotion?: Cartesia.Emotion[];
 }

package/api/resources/stt/client/Client.d.ts

@@ -1,8 +1,16 @@
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
 import * as environments from "../../../../environments";
 import * as core from "../../../../core";
+import * as Cartesia from "../../../index";
+import * as fs from "fs";
+import { Blob } from "buffer";
 export declare namespace Stt {
     interface Options {
         environment?: core.Supplier<environments.CartesiaEnvironment | string>;
+        /** Specify a custom URL to connect the client to. */
+        baseUrl?: core.Supplier<string>;
         apiKey?: core.Supplier<string | undefined>;
         /** Override the Cartesia-Version header */
         cartesiaVersion?: "2024-06-10";
@@ -17,11 +25,46 @@ export declare namespace Stt {
         abortSignal?: AbortSignal;
         /** Override the Cartesia-Version header */
         cartesiaVersion?: "2024-06-10";
+        /** Additional headers to include in the request. */
+        headers?: Record<string, string>;
     }
 }
 export declare class Stt {
     protected readonly _options: Stt.Options;
     constructor(_options?: Stt.Options);
+    /**
+     * Transcribes audio files into text using Cartesia's Speech-to-Text API.
+     *
+     * Upload an audio file and receive a complete transcription response. Supports arbitrarily long audio files with automatic intelligent chunking for longer audio.
+     *
+     * **Supported audio formats:** flac, m4a, mp3, mp4, mpeg, mpga, oga, ogg, wav, webm
+     *
+     * **Response format:** Returns JSON with transcribed text, duration, and language. Include `timestamp_granularities: ["word"]` to get word-level timestamps.
+     *
+     * **Pricing:** Batch transcription is priced at **1 credit per 2 seconds** of audio processed.
+     *
+     * <Note>
+     * For migrating from the OpenAI SDK, see our [OpenAI Whisper to Cartesia Ink Migration Guide](/api-reference/stt/migrate-from-open-ai).
+     * </Note>
+     *
+     * @param {File | fs.ReadStream | Blob} file
+     * @param {Cartesia.TranscriptionRequest} request
+     * @param {Stt.RequestOptions} requestOptions - Request-specific configuration.
+     *
+     * @example
+     *     await client.stt.transcribe(fs.createReadStream("/path/to/your/file"), {
+     *         model: "ink-whisper",
+     *         language: "en"
+     *     })
+     *
+     * @example
+     *     await client.stt.transcribe(fs.createReadStream("/path/to/your/file"), {
+     *         model: "ink-whisper",
+     *         language: "en",
+     *         timestampGranularities: ["word"]
+     *     })
+     */
+    transcribe(file: File | fs.ReadStream | Blob, request: Cartesia.TranscriptionRequest, requestOptions?: Stt.RequestOptions): Promise<Cartesia.TranscriptionResponse>;
     protected _getCustomAuthorizationHeaders(): Promise<{
         "X-API-Key": string | undefined;
     }>;

package/api/resources/stt/client/Client.js

@@ -1,4 +1,7 @@
 "use strict";
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
 var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
     if (k2 === undefined) k2 = k;
     var desc = Object.getOwnPropertyDescriptor(m, k);
@@ -41,13 +44,118 @@ var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, ge
         step((generator = generator.apply(thisArg, _arguments || [])).next());
     });
 };
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.Stt = void 0;
+const environments = __importStar(require("../../../../environments"));
 const core = __importStar(require("../../../../core"));
+const serializers = __importStar(require("../../../../serialization/index"));
+const url_join_1 = __importDefault(require("url-join"));
+const errors = __importStar(require("../../../../errors/index"));
 class Stt {
     constructor(_options = {}) {
         this._options = _options;
     }
+    /**
+     * Transcribes audio files into text using Cartesia's Speech-to-Text API.
+     *
+     * Upload an audio file and receive a complete transcription response. Supports arbitrarily long audio files with automatic intelligent chunking for longer audio.
+     *
+     * **Supported audio formats:** flac, m4a, mp3, mp4, mpeg, mpga, oga, ogg, wav, webm
+     *
+     * **Response format:** Returns JSON with transcribed text, duration, and language. Include `timestamp_granularities: ["word"]` to get word-level timestamps.
+     *
+     * **Pricing:** Batch transcription is priced at **1 credit per 2 seconds** of audio processed.
+     *
+     * <Note>
+     * For migrating from the OpenAI SDK, see our [OpenAI Whisper to Cartesia Ink Migration Guide](/api-reference/stt/migrate-from-open-ai).
+     * </Note>
+     *
+     * @param {File | fs.ReadStream | Blob} file
+     * @param {Cartesia.TranscriptionRequest} request
+     * @param {Stt.RequestOptions} requestOptions - Request-specific configuration.
+     *
+     * @example
+     *     await client.stt.transcribe(fs.createReadStream("/path/to/your/file"), {
+     *         model: "ink-whisper",
+     *         language: "en"
+     *     })
+     *
+     * @example
+     *     await client.stt.transcribe(fs.createReadStream("/path/to/your/file"), {
+     *         model: "ink-whisper",
+     *         language: "en",
+     *         timestampGranularities: ["word"]
+     *     })
+     */
+    transcribe(file, request, requestOptions) {
+        return __awaiter(this, void 0, void 0, function* () {
+            var _a, _b, _c, _d, _e, _f;
+            const _queryParams = {};
+            if (request.encoding != null) {
+                _queryParams["encoding"] = serializers.SttEncoding.jsonOrThrow(request.encoding, {
+                    unrecognizedObjectKeys: "strip",
+                });
+            }
+            if (request.sampleRate != null) {
+                _queryParams["sample_rate"] = request.sampleRate.toString();
+            }
+            const _request = yield core.newFormData();
+            yield _request.appendFile("file", file);
+            _request.append("model", request.model);
+            if (request.language != null) {
+                _request.append("language", request.language);
+            }
+            if (request.timestampGranularities != null) {
+                for (const _item of request.timestampGranularities) {
+                    _request.append("timestamp_granularities[]", serializers.TimestampGranularity.jsonOrThrow(_item, { unrecognizedObjectKeys: "strip" }));
+                }
+            }
+            const _maybeEncodedRequest = yield _request.getRequest();
+            const _response = yield ((_a = this._options.fetcher) !== null && _a !== void 0 ? _a : core.fetcher)({
+                url: (0, url_join_1.default)((_c = (_b = (yield core.Supplier.get(this._options.baseUrl))) !== null && _b !== void 0 ? _b : (yield core.Supplier.get(this._options.environment))) !== null && _c !== void 0 ? _c : environments.CartesiaEnvironment.Production, "/stt"),
+                method: "POST",
+                headers: Object.assign(Object.assign(Object.assign({ "Cartesia-Version": (_f = (_d = requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.cartesiaVersion) !== null && _d !== void 0 ? _d : (_e = this._options) === null || _e === void 0 ? void 0 : _e.cartesiaVersion) !== null && _f !== void 0 ? _f : "2024-06-10", "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "@cartesia/cartesia-js", "X-Fern-SDK-Version": "2.2.7", "User-Agent": "@cartesia/cartesia-js/2.2.7", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())), _maybeEncodedRequest.headers), requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.headers),
+                queryParameters: _queryParams,
+                requestType: "file",
+                duplex: _maybeEncodedRequest.duplex,
+                body: _maybeEncodedRequest.body,
+                timeoutMs: (requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) != null ? requestOptions.timeoutInSeconds * 1000 : 60000,
+                maxRetries: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.maxRetries,
+                abortSignal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
+            });
+            if (_response.ok) {
+                return serializers.TranscriptionResponse.parseOrThrow(_response.body, {
+                    unrecognizedObjectKeys: "passthrough",
+                    allowUnrecognizedUnionMembers: true,
+                    allowUnrecognizedEnumValues: true,
+                    skipValidation: true,
+                    breadcrumbsPrefix: ["response"],
+                });
+            }
+            if (_response.error.reason === "status-code") {
+                throw new errors.CartesiaError({
+                    statusCode: _response.error.statusCode,
+                    body: _response.error.body,
+                });
+            }
+            switch (_response.error.reason) {
+                case "non-json":
+                    throw new errors.CartesiaError({
+                        statusCode: _response.error.statusCode,
+                        body: _response.error.rawBody,
+                    });
+                case "timeout":
+                    throw new errors.CartesiaTimeoutError("Timeout exceeded when calling POST /stt.");
+                case "unknown":
+                    throw new errors.CartesiaError({
+                        message: _response.error.errorMessage,
+                    });
+            }
+        });
+    }
     _getCustomAuthorizationHeaders() {
         return __awaiter(this, void 0, void 0, function* () {
             const apiKeyValue = yield core.Supplier.get(this._options.apiKey);

package/api/resources/stt/client/index.d.ts

@@ -1 +1 @@
-export { Stt } from "./Client";
+export * from "./requests";

package/api/resources/stt/client/index.js

@@ -1,5 +1,17 @@
 "use strict";
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __exportStar = (this && this.__exportStar) || function(m, exports) {
+    for (var p in m) if (p !== "default" && !Object.prototype.hasOwnProperty.call(exports, p)) __createBinding(exports, m, p);
+};
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.Stt = void 0;
-var Client_1 = require("./Client");
-Object.defineProperty(exports, "Stt", { enumerable: true, get: function () { return Client_1.Stt; } });
+__exportStar(require("./requests"), exports);

package/api/resources/stt/client/requests/TranscriptionRequest.d.ts

@@ -0,0 +1,147 @@
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+import * as Cartesia from "../../../../index";
+/**
+ * @example
+ *     {
+ *         model: "ink-whisper",
+ *         language: "en"
+ *     }
+ *
+ * @example
+ *     {
+ *         model: "ink-whisper",
+ *         language: "en",
+ *         timestampGranularities: ["word"]
+ *     }
+ */
+export interface TranscriptionRequest {
+    /**
+     * The encoding format to process the audio as. If not specified, the audio file will be decoded automatically.
+     *
+     * **Supported formats:**
+     * - `pcm_s16le` - 16-bit signed integer PCM, little-endian (recommended for best performance)
+     * - `pcm_s32le` - 32-bit signed integer PCM, little-endian
+     * - `pcm_f16le` - 16-bit floating point PCM, little-endian
+     * - `pcm_f32le` - 32-bit floating point PCM, little-endian
+     * - `pcm_mulaw` - 8-bit μ-law encoded PCM
+     * - `pcm_alaw` - 8-bit A-law encoded PCM
+     */
+    encoding?: Cartesia.SttEncoding;
+    /**
+     * The sample rate of the audio in Hz.
+     */
+    sampleRate?: number;
+    /** ID of the model to use for transcription. Use `ink-whisper` for the latest Cartesia Whisper model. */
+    model: string;
+    /**
+     * The language of the input audio in ISO-639-1 format. Defaults to `en`.
+     *
+     * <Accordion title="Supported languages">
+     *   - `en` (English)
+     *   - `zh` (Chinese)
+     *   - `de` (German)
+     *   - `es` (Spanish)
+     *   - `ru` (Russian)
+     *   - `ko` (Korean)
+     *   - `fr` (French)
+     *   - `ja` (Japanese)
+     *   - `pt` (Portuguese)
+     *   - `tr` (Turkish)
+     *   - `pl` (Polish)
+     *   - `ca` (Catalan)
+     *   - `nl` (Dutch)
+     *   - `ar` (Arabic)
+     *   - `sv` (Swedish)
+     *   - `it` (Italian)
+     *   - `id` (Indonesian)
+     *   - `hi` (Hindi)
+     *   - `fi` (Finnish)
+     *   - `vi` (Vietnamese)
+     *   - `he` (Hebrew)
+     *   - `uk` (Ukrainian)
+     *   - `el` (Greek)
+     *   - `ms` (Malay)
+     *   - `cs` (Czech)
+     *   - `ro` (Romanian)
+     *   - `da` (Danish)
+     *   - `hu` (Hungarian)
+     *   - `ta` (Tamil)
+     *   - `no` (Norwegian)
+     *   - `th` (Thai)
+     *   - `ur` (Urdu)
+     *   - `hr` (Croatian)
+     *   - `bg` (Bulgarian)
+     *   - `lt` (Lithuanian)
+     *   - `la` (Latin)
+     *   - `mi` (Maori)
+     *   - `ml` (Malayalam)
+     *   - `cy` (Welsh)
+     *   - `sk` (Slovak)
+     *   - `te` (Telugu)
+     *   - `fa` (Persian)
+     *   - `lv` (Latvian)
+     *   - `bn` (Bengali)
+     *   - `sr` (Serbian)
+     *   - `az` (Azerbaijani)
+     *   - `sl` (Slovenian)
+     *   - `kn` (Kannada)
+     *   - `et` (Estonian)
+     *   - `mk` (Macedonian)
+     *   - `br` (Breton)
+     *   - `eu` (Basque)
+     *   - `is` (Icelandic)
+     *   - `hy` (Armenian)
+     *   - `ne` (Nepali)
+     *   - `mn` (Mongolian)
+     *   - `bs` (Bosnian)
+     *   - `kk` (Kazakh)
+     *   - `sq` (Albanian)
+     *   - `sw` (Swahili)
+     *   - `gl` (Galician)
+     *   - `mr` (Marathi)
+     *   - `pa` (Punjabi)
+     *   - `si` (Sinhala)
+     *   - `km` (Khmer)
+     *   - `sn` (Shona)
+     *   - `yo` (Yoruba)
+     *   - `so` (Somali)
+     *   - `af` (Afrikaans)
+     *   - `oc` (Occitan)
+     *   - `ka` (Georgian)
+     *   - `be` (Belarusian)
+     *   - `tg` (Tajik)
+     *   - `sd` (Sindhi)
+     *   - `gu` (Gujarati)
+     *   - `am` (Amharic)
+     *   - `yi` (Yiddish)
+     *   - `lo` (Lao)
+     *   - `uz` (Uzbek)
+     *   - `fo` (Faroese)
+     *   - `ht` (Haitian Creole)
+     *   - `ps` (Pashto)
+     *   - `tk` (Turkmen)
+     *   - `nn` (Nynorsk)
+     *   - `mt` (Maltese)
+     *   - `sa` (Sanskrit)
+     *   - `lb` (Luxembourgish)
+     *   - `my` (Myanmar)
+     *   - `bo` (Tibetan)
+     *   - `tl` (Tagalog)
+     *   - `mg` (Malagasy)
+     *   - `as` (Assamese)
+     *   - `tt` (Tatar)
+     *   - `haw` (Hawaiian)
+     *   - `ln` (Lingala)
+     *   - `ha` (Hausa)
+     *   - `ba` (Bashkir)
+     *   - `jw` (Javanese)
+     *   - `su` (Sundanese)
+     *   - `yue` (Cantonese)
+     * </Accordion>
+     */
+    language?: string;
+    /** The timestamp granularities to populate for this transcription. Currently only `word` level timestamps are supported. */
+    timestampGranularities?: Cartesia.TimestampGranularity[];
+}

package/api/resources/stt/client/requests/TranscriptionRequest.js

@@ -0,0 +1,5 @@
+"use strict";
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+Object.defineProperty(exports, "__esModule", { value: true });

package/api/resources/stt/client/requests/index.d.ts

@@ -0,0 +1 @@
+export { type TranscriptionRequest } from "./TranscriptionRequest";

package/api/resources/stt/client/requests/index.js

@@ -0,0 +1,2 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });

package/api/resources/stt/index.d.ts

@@ -1 +1,2 @@
 export * from "./types";
+export * from "./client";

package/api/resources/stt/index.js

@@ -15,3 +15,4 @@ var __exportStar = (this && this.__exportStar) || function(m, exports) {
 };
 Object.defineProperty(exports, "__esModule", { value: true });
 __exportStar(require("./types"), exports);
+__exportStar(require("./client"), exports);

package/api/resources/stt/types/TimestampGranularity.d.ts

@@ -0,0 +1,12 @@
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+/**
+ * The granularity of timestamps to include in the response.
+ *
+ * Currently only `word` level timestamps are supported, providing start and end times for each word.
+ */
+export type TimestampGranularity = "word";
+export declare const TimestampGranularity: {
+    readonly Word: "word";
+};

package/api/resources/stt/types/TimestampGranularity.js

@@ -0,0 +1,9 @@
+"use strict";
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.TimestampGranularity = void 0;
+exports.TimestampGranularity = {
+    Word: "word",
+};

package/api/resources/stt/types/TranscriptMessage.d.ts

@@ -15,7 +15,7 @@ export interface TranscriptMessage {
     isFinal: boolean;
     /** The duration of the audio transcribed so far, in seconds. */
     duration?: number;
-    /** The detected or specified language of the input audio. */
+    /** The specified language of the input audio. */
     language?: string;
     /** Word-level timestamps showing the start and end time of each word in seconds. Always included in streaming responses. */
     words?: Cartesia.TranscriptionWord[];

package/api/resources/stt/types/TranscriptionResponse.d.ts

@@ -5,10 +5,10 @@ import * as Cartesia from "../../../index";
 export interface TranscriptionResponse {
     /** The transcribed text. */
     text: string;
-    /** The detected or specified language of the input audio. */
+    /** The specified language of the input audio. */
     language?: string;
     /** The duration of the input audio in seconds. */
     duration?: number;
-    /** Word-level timestamps for batch transcription responses. */
+    /** Word-level timestamps showing the start and end time of each word. Only included when `[word]` is passed into `timestamp_granularities[]`. */
     words?: Cartesia.TranscriptionWord[];
 }

package/api/resources/stt/types/index.d.ts

@@ -1,3 +1,4 @@
+export * from "./TimestampGranularity";
 export * from "./TranscriptionWord";
 export * from "./TranscriptionResponse";
 export * from "./StreamingTranscriptionResponse";

package/api/resources/stt/types/index.js

@@ -14,6 +14,7 @@ var __exportStar = (this && this.__exportStar) || function(m, exports) {
     for (var p in m) if (p !== "default" && !Object.prototype.hasOwnProperty.call(exports, p)) __createBinding(exports, m, p);
 };
 Object.defineProperty(exports, "__esModule", { value: true });
+__exportStar(require("./TimestampGranularity"), exports);
 __exportStar(require("./TranscriptionWord"), exports);
 __exportStar(require("./TranscriptionResponse"), exports);
 __exportStar(require("./StreamingTranscriptionResponse"), exports);

package/api/resources/tts/client/Client.d.ts

@@ -4,9 +4,12 @@
 import * as environments from "../../../../environments";
 import * as core from "../../../../core";
 import * as Cartesia from "../../../index";
+import * as stream from "stream";
 export declare namespace Tts {
     interface Options {
         environment?: core.Supplier<environments.CartesiaEnvironment | string>;
+        /** Specify a custom URL to connect the client to. */
+        baseUrl?: core.Supplier<string>;
         apiKey?: core.Supplier<string | undefined>;
         /** Override the Cartesia-Version header */
         cartesiaVersion?: "2024-06-10";
@@ -21,13 +24,15 @@ export declare namespace Tts {
         abortSignal?: AbortSignal;
         /** Override the Cartesia-Version header */
         cartesiaVersion?: "2024-06-10";
+        /** Additional headers to include in the request. */
+        headers?: Record<string, string>;
     }
 }
 export declare class Tts {
     protected readonly _options: Tts.Options;
     constructor(_options?: Tts.Options);
-    bytes(request: Cartesia.TtsRequest, requestOptions?: Tts.RequestOptions): Promise<ArrayBuffer>;
-    sse(request: Cartesia.TtsRequest, requestOptions?: Tts.RequestOptions): Promise<core.Stream<Cartesia.WebSocketResponse>>;
+    bytes(request: Cartesia.TtsRequest, requestOptions?: Tts.RequestOptions): Promise<stream.Readable>;
+    sse(request: Cartesia.TtssseRequest, requestOptions?: Tts.RequestOptions): Promise<core.Stream<Cartesia.WebSocketResponse>>;
     protected _getCustomAuthorizationHeaders(): Promise<{
         "X-API-Key": string | undefined;
     }>;