js-tts-wrapper/cjs/engines/azure.d.ts

A JavaScript/TypeScript library that provides a unified API for working with multiple cloud-based Text-to-Speech (TTS) services

import { AbstractTTSClient } from "../core/abstract-tts";
import type { SpeakOptions, TTSCredentials, UnifiedVoice, WordBoundaryCallback } from "../types";
/**
 * Azure TTS Client Credentials
 */
export interface AzureTTSCredentials extends TTSCredentials {
    subscriptionKey: string;
    region: string;
}
/**
 * Azure TTS Client
 */
export declare class AzureTTSClient extends AbstractTTSClient {
    private subscriptionKey;
    private region;
    private sdk;
    private sdkLoadingPromise;
    /**
     * Create a new Azure TTS client
     * @param credentials Azure credentials object with subscriptionKey and region
     */
    constructor(credentials: AzureTTSCredentials);
    /**
     * Get raw voices from Azure
     * @returns Promise resolving to an array of unified voice objects
     */
    protected _getVoices(): Promise<any[]>;
    /**
     * Map Azure voice objects to unified format
     * @param rawVoices Array of Azure voice objects
     * @returns Promise resolving to an array of unified voice objects
     */
    protected _mapVoicesToUnified(rawVoices: any[]): Promise<UnifiedVoice[]>;
    /**
     * Synthesize text to audio bytes
     * @param text Text or SSML to synthesize
     * @param options Synthesis options
     * @returns Promise resolving to audio bytes
     */
    synthToBytes(text: string, options?: AzureTTSOptions): Promise<Uint8Array>;
    /**
     * Synthesize text to a byte stream with word boundary information
     * @param text Text or SSML to synthesize
     * @param options Synthesis options
     * @returns Promise resolving to an object containing the audio stream and word boundary information
     */
    synthToBytestream(text: string, options?: AzureTTSOptions): Promise<{
        audioStream: ReadableStream<Uint8Array>;
        wordBoundaries: Array<{
            text: string;
            offset: number;
            duration: number;
        }>;
    }>;
    /**
     * Load the Microsoft Speech SDK dynamically.
     * @returns A promise resolving to the SDK module, or null if loading fails or not applicable.
     */
    private loadSDK;
    /**
     * Synthesize speech using the Microsoft Cognitive Services Speech SDK
     * @param ssml SSML to synthesize
     * @param options Synthesis options
     * @param sdkInstance The loaded SDK instance.
     * @returns Promise resolving to an object containing the audio stream and word boundary information
     */
    private synthToBytestreamWithSDK;
    /**
     * Synthesize speech using the REST API (no word boundaries)
     * @param ssml SSML to synthesize
     * @param options Synthesis options
     * @returns Promise resolving to an object containing the audio stream and an empty word boundary array
     */
    private synthToBytestreamWithREST;
    /**
     * Start playback with word boundary callbacks
     * @param text Text or SSML to speak
     * @param callback Callback function for word boundaries
     * @param options Synthesis options
     */
    startPlaybackWithCallbacks(text: string, callback: WordBoundaryCallback, options?: AzureTTSOptions): Promise<void>;
    /**
     * Start playback with word boundary callbacks using the SDK
     * @param text Text or SSML to speak
     * @param callback Callback function for word boundaries
     * @param options Synthesis options
     */
    private startPlaybackWithCallbacksSDK;
    /**
     * Prepare SSML for synthesis
     * @param text Text or SSML to prepare
     * @param options Synthesis options
     * @returns SSML ready for synthesis
     */
    private prepareSSML;
}
/**
 * Extended options for Azure TTS
 */
export interface AzureTTSOptions extends SpeakOptions {
    format?: 'mp3' | 'wav';
}