import { Buffer } from 'buffer';
import { APIResource } from "../resource.js";
export interface Timestamp {
    /**
     * The text segment
     */
    text: string;
    /**
     * The time at which the text starts, in seconds
     */
    start: number;
    /**
     * The spoken duration of the text segment, in seconds
     */
    duration: number;
}
export interface ErrorBody {
    /**
     * Slug identifying the error category.
     */
    type: 'invalid_request_error' | 'authentication_error' | 'permission_error' | 'not_found_error' | 'payment_required_error' | 'rate_limit_error' | 'internal_server_error';
    /**
     * Human-readable error message.
     */
    message: string;
}
export type SpeechSessionMessage = SpeechSessionAudio | SpeechSessionReady | SpeechSessionTimestamps | SpeechSessionFlushComplete | SpeechSessionResetComplete | SpeechSessionError;
/**
 * Binary audio data returned from the server
 */
export interface SpeechSessionAudio {
    type: 'audio';
    audio: ArrayBuffer | Buffer;
}
/**
 * First message sent by the server, confirming the session is established.
 */
export interface SpeechSessionReady {
    type: 'ready';
    request_id: string;
}
/**
 * Timestamps for the audio chunk that was just streamed, if requested in `init`.
 */
export interface SpeechSessionTimestamps {
    type: 'timestamps';
    timestamps?: Array<Timestamp>;
}
/**
 * Acknowledgement that a flush command has been completed.
 *
 * The `nonce` matches the one carried by the original flush, allowing you to determine when it has completed.
 */
export interface SpeechSessionFlushComplete {
    type: 'flush_complete';
    nonce: number;
}
/**
 * Acknowledgement that a reset command has been completed.
 *
 * The `nonce` matches the one carried by the original reset, allowing you to discard any remaining in-flight speech before the reset.
 */
export interface SpeechSessionResetComplete {
    type: 'reset_complete';
    nonce: number;
}
/**
 * Error envelope returned by the server. Connection closes immediately afterward.
 */
export interface SpeechSessionError {
    type: 'error';
    error: ErrorBody;
    request_id: string;
}
export declare class SpeechSession {
    private socket;
    private outMessages;
    private inMessages;
    private nextNonce;
    /** Per-session request id, populated when the server emits its `ready` frame. */
    request_id: string | null;
    constructor(apiKey: string, options: SpeechSessionParams, baseURL?: string);
    private setupWebSocket;
    private onError;
    private onOpen;
    private onClose;
    private onMessage;
    /**
     * Close the SpeechSession.
     */
    close(): void;
    /**
     * Send text to the server to append into the text stream.
     */
    sendText(text: string): void;
    /**
     * Force the server to generate speech for all buffered text in the stream.
     *
     * @returns The nonce used for this flush command
     */
    sendFlush(): number;
    /**
     * Drop the server's buffered text without generating speech for it.
     *
     * @returns The nonce used for this reset command
     */
    sendReset(): number;
    /**
     * Inform the server you're done appending text to this session and want it to close when the server has finished dispatching speech.
     */
    sendFinish(): void;
    /**
     * Iterate over messages from the server.
     *
     * @returns AsyncIterator<SpeechSessionMessage>
     */
    [Symbol.asyncIterator](): AsyncIterator<SpeechSessionMessage>;
    streamAudioToFile(path: string): Promise<void>;
    private parseTextMessage;
    private sendMessage;
    private flushMessages;
}
export declare class Sessions extends APIResource {
    /**
     * Stream text to our servers and receive generated speech in real-time. Great for latency-sensitive applications and situations where you don't have all the text upfront.
     */
    create(body: SpeechSessionParams): SpeechSession;
}
export interface SpeechSessionParams {
    /**
     * The voice ID to use for speech generation, obtained from 'List voices' API
     */
    voice: string;
    /**
     * The desired output format of the audio.
     * - `mp3`: 96kbps MP3 audio.
     * - `pcm_s16le`: PCM signed 16-bit little-endian audio.
     * - `pcm_f32le`: PCM 32-bit floating-point little-endian audio.
     * - `ulaw`: 8-bit G711 µ-law audio with a WAV header.
     * - `webm`: WebM format with Opus audio codec.
     */
    format?: 'mp3' | 'pcm_s16le' | 'pcm_f32le' | 'ulaw' | 'webm';
    language?: 'auto' | 'ar' | 'as' | 'bn' | 'cs' | 'da' | 'de' | 'en' | 'es' | 'fi' | 'fr' | 'hi' | 'id' | 'it' | 'ja' | 'ko' | 'ml' | 'mr' | 'nl' | 'pl' | 'pt' | 'ru' | 'sk' | 'sv' | 'ta' | 'te' | 'th' | 'tr' | 'uk' | 'ur' | 'vi' | 'zh';
    /**
     * Controls whether the server will return timestamps for the generated speech
     */
    return_timestamps?: boolean;
    /**
     * The desired output audio sample rate
     */
    sample_rate?: 24000 | 16000 | 8000;
}
export declare namespace Sessions {
    export { type SpeechSessionParams as SpeechSessionParams };
    export { SpeechSession as SpeechSession };
}
//# sourceMappingURL=sessions.d.ts.map