import type { VoiceModeCapability } from '@xopcai/realtime-protocol/voice';
export declare function platformVoiceModels(mode: VoiceModeCapability): import("../providers/model-catalog-types.js").CatalogModel[];
export declare function requirePlatformVoiceModel(id: string, mode: VoiceModeCapability): {
    voice: {
        protocolVersion: 1;
        serviceVersion: number;
        modes: ("conversation" | "transcription" | "speech" | "transcription.stream" | "speech.stream")[];
        transport: "websocket-pcm";
        inputFormat: {
            encoding: "pcm_s16le";
            sampleRate: 16000;
            channels: 1;
        };
        outputFormat: {
            encoding: "pcm_s16le";
            sampleRate: 24000;
            channels: 1;
        };
        turnDetection: "server_vad"[];
        bargeIn: boolean;
        tools: boolean;
        resumable: false;
        voices: {
            id: string;
            name: string;
            languages: string[];
            gender?: "female" | "male";
            style?: string;
            description?: string;
        }[];
        limits: {
            maxFrameBytes: number;
            maxSessionSeconds: number;
        };
        defaultVoice?: string;
    };
    computerUse?: {
        profile: import("@xopcai/computer-control-contract/src/index.js").ComputerProfile;
    };
    id: string;
    name: string;
    displayNames?: Partial<Record<"zh-CN" | "en", string>>;
    availability: "available" | "unavailable";
    kind: "language" | "image" | "stt" | "tts" | "omni";
    input: Array<"text" | "image" | "audio">;
    output: Array<"text" | "image" | "audio">;
    operations: Array<"chat.completions" | "responses" | "images.generate" | "images.edit" | "audio.transcription" | "audio.speech" | "audio.conversation">;
    reasoning: boolean;
    contextWindow: number;
    maxOutputTokens: number | null;
    stability?: "stable" | "preview" | "deprecated";
    priority?: number;
    tier?: string;
    bestEffort?: boolean;
    imageGeneration?: {
        maxCount: number;
        sizes: string[];
        aspectRatios?: string[];
        qualities: Array<"low" | "medium" | "high" | "auto">;
        formats: Array<"png" | "jpeg" | "webp">;
        backgrounds: Array<"transparent" | "opaque" | "auto">;
        maxInputImages: number;
    };
    stt?: {
        inputFormats: string[];
        maxBytes: number;
        maxDurationSeconds: number;
        languages: string[];
        languageHint: boolean;
        prompt: boolean;
        timestamps: Array<"segment" | "word">;
        diarization: boolean;
    };
    tts?: {
        maxCharacters: number;
        languages: string[];
        outputFormats: Array<"mp3" | "opus" | "aac" | "flac" | "wav" | "pcm">;
        streaming: boolean;
        speed: boolean;
        pitch: boolean;
        instructions: boolean;
        defaultVoice?: string;
    };
};