@cartesia/cartesia-js 2.2.2 → 2.2.4

package/README.md

@@ -69,13 +69,6 @@ const websocket = cartesia.tts.websocket({
     sampleRate: 44100,
 });
 
-try {
-    await websocket.connect();
-} catch (error) {
-    console.error(`Failed to connect to Cartesia: ${error}`);
-    throw error;
-}
-
 // Create a stream.
 const response = await websocket.send({
     modelId: "sonic-2",
@@ -148,6 +141,64 @@ await player.play(response.source);
 console.log("Done playing.");
 ```
 
+## Speech-to-Text (STT)
+
+```typescript
+import { CartesiaClient } from "@cartesia/cartesia-js";
+import fs from "fs";
+
+const client = new CartesiaClient({
+    apiKey: process.env.CARTESIA_API_KEY,
+});
+
+// Create STT WebSocket connection
+const sttWs = client.stt.websocket({
+    model: "ink-whisper",
+    language: "en",
+    encoding: "pcm_s16le",
+    sampleRate: 16000,
+});
+
+// Set up message handler
+await sttWs.onMessage((result) => {
+    if (result.type === "transcript") {
+        const status = result.isFinal ? "FINAL" : "INTERIM";
+        console.log(`[${status}] ${result.text}`);
+        if (result.duration) {
+            console.log(`Duration: ${result.duration.toFixed(2)}s`);
+        }
+    } else if (result.type === "flush_done") {
+        console.log("Flush completed");
+        await sttWs.done(); // Send done command
+    } else if (result.type === "done") {
+        console.log("Session complete");
+    } else if (result.type === "error") {
+        console.error(`Error: ${result.message}`);
+    }
+});
+
+// Load and send audio data
+const audioBuffer = fs.readFileSync("audio.wav");
+const chunkSize = 1600; // ~100ms at 16kHz
+const audioChunks = [];
+
+for (let i = 0; i < audioBuffer.length; i += chunkSize) {
+    const chunk = audioBuffer.slice(i, i + chunkSize);
+    audioChunks.push(chunk.buffer);
+}
+
+// Send audio chunks
+for (const chunk of audioChunks) {
+    await sttWs.send(chunk);
+}
+
+// Finalize transcription
+await sttWs.finalize();
+
+// Disconnect when done
+sttWs.disconnect();
+```
+
 ## Request And Response Types
 
 The SDK exports all request and response types as TypeScript interfaces. Simply import them with the

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": "2.2.2", "User-Agent": "@cartesia/cartesia-js/2.2.2", "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": "2.2.4", "User-Agent": "@cartesia/cartesia-js/2.2.4", "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.js

@@ -78,7 +78,7 @@ 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": "2.2.2", "User-Agent": "@cartesia/cartesia-js/2.2.2", "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": "2.2.4", "User-Agent": "@cartesia/cartesia-js/2.2.4", "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/index.d.ts

@@ -4,6 +4,8 @@ export * as auth from "./auth";
 export * from "./auth/types";
 export * as embedding from "./embedding";
 export * from "./embedding/types";
+export * as stt from "./stt";
+export * from "./stt/types";
 export * as tts from "./tts";
 export * from "./tts/types";
 export * as voiceChanger from "./voiceChanger";

package/api/resources/index.js

@@ -36,13 +36,15 @@ 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.infill = exports.voices = exports.voiceChanger = exports.tts = exports.embedding = exports.auth = exports.apiStatus = void 0;
+exports.infill = exports.voices = exports.voiceChanger = exports.tts = exports.stt = exports.embedding = exports.auth = exports.apiStatus = void 0;
 exports.apiStatus = __importStar(require("./apiStatus"));
 __exportStar(require("./apiStatus/types"), exports);
 exports.auth = __importStar(require("./auth"));
 __exportStar(require("./auth/types"), exports);
 exports.embedding = __importStar(require("./embedding"));
 __exportStar(require("./embedding/types"), exports);
+exports.stt = __importStar(require("./stt"));
+__exportStar(require("./stt/types"), exports);
 exports.tts = __importStar(require("./tts"));
 __exportStar(require("./tts/types"), exports);
 exports.voiceChanger = __importStar(require("./voiceChanger"));

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": "2.2.2", "User-Agent": "@cartesia/cartesia-js/2.2.2", "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": "2.2.4", "User-Agent": "@cartesia/cartesia-js/2.2.4", "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/stt/client/Client.d.ts

@@ -0,0 +1,28 @@
+import * as environments from "../../../../environments";
+import * as core from "../../../../core";
+export declare namespace Stt {
+    interface Options {
+        environment?: core.Supplier<environments.CartesiaEnvironment | string>;
+        apiKey?: core.Supplier<string | undefined>;
+        /** Override the Cartesia-Version header */
+        cartesiaVersion?: "2024-06-10";
+        fetcher?: core.FetchFunction;
+    }
+    interface RequestOptions {
+        /** The maximum time to wait for a response in seconds. */
+        timeoutInSeconds?: number;
+        /** The number of times to retry the request. Defaults to 2. */
+        maxRetries?: number;
+        /** A hook to abort the request. */
+        abortSignal?: AbortSignal;
+        /** Override the Cartesia-Version header */
+        cartesiaVersion?: "2024-06-10";
+    }
+}
+export declare class Stt {
+    protected readonly _options: Stt.Options;
+    constructor(_options?: Stt.Options);
+    protected _getCustomAuthorizationHeaders(): Promise<{
+        "X-API-Key": string | undefined;
+    }>;
+}

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

@@ -0,0 +1,58 @@
+"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 __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
+    function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
+    return new (P || (P = Promise))(function (resolve, reject) {
+        function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
+        function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
+        function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
+        step((generator = generator.apply(thisArg, _arguments || [])).next());
+    });
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.Stt = void 0;
+const core = __importStar(require("../../../../core"));
+class Stt {
+    constructor(_options = {}) {
+        this._options = _options;
+    }
+    _getCustomAuthorizationHeaders() {
+        return __awaiter(this, void 0, void 0, function* () {
+            const apiKeyValue = yield core.Supplier.get(this._options.apiKey);
+            return { "X-API-Key": apiKeyValue };
+        });
+    }
+}
+exports.Stt = Stt;

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

@@ -0,0 +1 @@
+export { Stt } from "./Client";

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

@@ -0,0 +1,5 @@
+"use strict";
+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; } });

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

@@ -0,0 +1 @@
+export * from "./types";

package/api/resources/stt/index.js

@@ -0,0 +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 });
+__exportStar(require("./types"), exports);

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

@@ -0,0 +1,10 @@
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+/**
+ * Acknowledgment message sent in response to a `done` command, indicating that the session is complete and the WebSocket will close.
+ */
+export interface DoneMessage {
+    /** Unique identifier for this transcription session. */
+    requestId: string;
+}

package/api/resources/stt/types/DoneMessage.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/types/ErrorMessage.d.ts

@@ -0,0 +1,9 @@
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+export interface ErrorMessage {
+    /** The request ID associated with the error, if applicable. */
+    requestId?: string;
+    /** Human-readable error message describing what went wrong. */
+    message: string;
+}

package/api/resources/stt/types/ErrorMessage.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/types/FlushDoneMessage.d.ts

@@ -0,0 +1,10 @@
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+/**
+ * Acknowledgment message sent in response to a `finalize` command, indicating that all buffered audio has been flushed and processed.
+ */
+export interface FlushDoneMessage {
+    /** Unique identifier for this transcription session. */
+    requestId: string;
+}

package/api/resources/stt/types/FlushDoneMessage.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/types/StreamingTranscriptionResponse.d.ts

@@ -0,0 +1,22 @@
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+import * as Cartesia from "../../../index";
+/**
+ * The server sends transcription results, control messages, or error messages. Each message has a `type` field to distinguish between different message types.
+ */
+export type StreamingTranscriptionResponse = Cartesia.StreamingTranscriptionResponse.Transcript | Cartesia.StreamingTranscriptionResponse.FlushDone | Cartesia.StreamingTranscriptionResponse.Done | Cartesia.StreamingTranscriptionResponse.Error_;
+export declare namespace StreamingTranscriptionResponse {
+    interface Transcript extends Cartesia.TranscriptMessage {
+        type: "transcript";
+    }
+    interface FlushDone extends Cartesia.FlushDoneMessage {
+        type: "flush_done";
+    }
+    interface Done extends Cartesia.DoneMessage {
+        type: "done";
+    }
+    interface Error_ extends Cartesia.ErrorMessage {
+        type: "error";
+    }
+}

package/api/resources/stt/types/StreamingTranscriptionResponse.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/types/SttEncoding.d.ts

@@ -0,0 +1,15 @@
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+/**
+ * The encoding format for audio data sent to the STT WebSocket.
+ *
+ * Currently supported:
+ * - `pcm_s16le` - 16-bit signed integer PCM, little-endian
+ *
+ * Support for other formats will be added in the future.
+ */
+export type SttEncoding = "pcm_s16le";
+export declare const SttEncoding: {
+    readonly PcmS16Le: "pcm_s16le";
+};

package/api/resources/stt/types/SttEncoding.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.SttEncoding = void 0;
+exports.SttEncoding = {
+    PcmS16Le: "pcm_s16le",
+};

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

@@ -0,0 +1,19 @@
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+export interface TranscriptMessage {
+    /** Unique identifier for this transcription session. */
+    requestId: string;
+    /**
+     * The transcribed text. May be partial or final depending on is_final.
+     *
+     * **Note**: Text may be empty in initial responses while the system accumulates sufficient audio for transcription. This is normal behavior - wait for responses with non-empty text or monitor is_final for completion status.
+     */
+    text: string;
+    /** Whether this is a final transcription result or an interim result. */
+    isFinal: boolean;
+    /** The duration of the audio transcribed so far, in seconds. */
+    duration?: number;
+    /** The detected or specified language of the input audio. */
+    language?: string;
+}

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

@@ -0,0 +1,11 @@
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+export interface TranscriptionResponse {
+    /** The transcribed text. */
+    text: string;
+    /** The detected or specified language of the input audio. */
+    language?: string;
+    /** The duration of the input audio in seconds. */
+    duration?: number;
+}

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

@@ -0,0 +1,7 @@
+export * from "./TranscriptionResponse";
+export * from "./StreamingTranscriptionResponse";
+export * from "./TranscriptMessage";
+export * from "./FlushDoneMessage";
+export * from "./DoneMessage";
+export * from "./ErrorMessage";
+export * from "./SttEncoding";

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

@@ -0,0 +1,23 @@
+"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 });
+__exportStar(require("./TranscriptionResponse"), exports);
+__exportStar(require("./StreamingTranscriptionResponse"), exports);
+__exportStar(require("./TranscriptMessage"), exports);
+__exportStar(require("./FlushDoneMessage"), exports);
+__exportStar(require("./DoneMessage"), exports);
+__exportStar(require("./ErrorMessage"), exports);
+__exportStar(require("./SttEncoding"), exports);

package/api/resources/tts/types/GenerationRequest.d.ts

@@ -15,14 +15,7 @@ export interface GenerationRequest {
      * If the duration is not appropriate for the length of the transcript, the output audio may be truncated.
      */
     duration?: number;
-    /**
-     * The text [classifier-free guidance](https://arxiv.org/abs/2207.12598) value for the request.
-     *
-     * Higher values causes the model to attend more to the text but speed up the generation. Lower values reduce the speaking rate but can increase the risk of hallucinations. The default value is `3.0`. For a slower speaking rate, we recommend values between `2.0` and `3.0`. Values are supported between `1.5` and `3.0`.
-     *
-     * This parameter is only supported for `sonic-2` models.
-     */
-    textCfg?: number;
+    speed?: Cartesia.ModelSpeed;
     contextId?: Cartesia.ContextId;
     /**
      * Whether this input may be followed by more inputs.

package/api/resources/tts/types/ModelSpeed.d.ts

@@ -0,0 +1,16 @@
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+/**
+ * > This feature is experimental and may not work for all voices.
+ *
+ * Speed setting for the model. Defaults to `normal`.
+ *
+ * Influences the speed of the generated speech. Faster speeds may reduce hallucination rate.
+ */
+export type ModelSpeed = "slow" | "normal" | "fast";
+export declare const ModelSpeed: {
+    readonly Slow: "slow";
+    readonly Normal: "normal";
+    readonly Fast: "fast";
+};

package/api/resources/tts/types/ModelSpeed.js

@@ -0,0 +1,11 @@
+"use strict";
+/**
+ * This file was auto-generated by Fern from our API Definition.
+ */
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.ModelSpeed = void 0;
+exports.ModelSpeed = {
+    Slow: "slow",
+    Normal: "normal",
+    Fast: "fast",
+};

package/api/resources/tts/types/TtsRequest.d.ts

@@ -14,12 +14,5 @@ export interface TtsRequest {
      * If the duration is not appropriate for the length of the transcript, the output audio may be truncated.
      */
     duration?: number;
-    /**
-     * The text [classifier-free guidance](https://arxiv.org/abs/2207.12598) value for the request.
-     *
-     * Higher values causes the model to attend more to the text but speed up the generation. Lower values reduce the speaking rate but can increase the risk of hallucinations. The default value is `3.0`. For a slower speaking rate, we recommend values between `2.0` and `3.0`. Values are supported between `1.5` and `3.0`.
-     *
-     * This parameter is only supported for `sonic-2` models.
-     */
-    textCfg?: number;
+    speed?: Cartesia.ModelSpeed;
 }

package/api/resources/tts/types/WebSocketTtsRequest.d.ts

@@ -16,5 +16,5 @@ export interface WebSocketTtsRequest {
     continue?: boolean;
     contextId?: string;
     maxBufferDelayMs?: number;
-    textCfg?: number;
+    speed?: Cartesia.ModelSpeed;
 }

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

@@ -1,5 +1,6 @@
 export * from "./ContextId";
 export * from "./FlushId";
+export * from "./ModelSpeed";
 export * from "./WebSocketBaseResponse";
 export * from "./WebSocketResponse";
 export * from "./WebSocketErrorResponse";

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

@@ -16,6 +16,7 @@ var __exportStar = (this && this.__exportStar) || function(m, exports) {
 Object.defineProperty(exports, "__esModule", { value: true });
 __exportStar(require("./ContextId"), exports);
 __exportStar(require("./FlushId"), exports);
+__exportStar(require("./ModelSpeed"), exports);
 __exportStar(require("./WebSocketBaseResponse"), exports);
 __exportStar(require("./WebSocketResponse"), exports);
 __exportStar(require("./WebSocketErrorResponse"), exports);

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

@@ -55,8 +55,7 @@ export declare class Voices {
      *         name: "A high-stability cloned voice",
      *         description: "Copied from Cartesia docs",
      *         mode: "stability",
-     *         language: "en",
-     *         enhance: true
+     *         language: "en"
      *     })
      *
      * @example
@@ -64,9 +63,7 @@ export declare class Voices {
      *         name: "A high-similarity cloned voice",
      *         description: "Copied from Cartesia docs",
      *         mode: "similarity",
-     *         language: "en",
-     *         transcript: "A transcript of the words spoken in the audio clip.",
-     *         enhance: false
+     *         language: "en"
      *     })
      */
     clone(clip: File | fs.ReadStream | Blob, request: Cartesia.CloneVoiceRequest, requestOptions?: Voices.RequestOptions): Promise<Cartesia.VoiceMetadata>;