@chinchillaenterprises/mcp-ai-assistant/dist/tools/tts/text-to-speech.js

MCP server for AI assistant with ElevenLabs text-to-speech integration

"use strict";
var __importDefault = (this && this.__importDefault) || function (mod) {
    return (mod && mod.__esModule) ? mod : { "default": mod };
};
Object.defineProperty(exports, "__esModule", { value: true });
exports.textToSpeechTool = void 0;
const zod_1 = require("zod");
const fs_1 = require("fs");
const path_1 = __importDefault(require("path"));
const account_state_js_1 = require("../../services/account-state.js");
const elevenlabs_client_js_1 = require("../../services/elevenlabs-client.js");
const inputSchema = zod_1.z.object({
    text: zod_1.z.string().describe('The text to convert to speech'),
    voiceId: zod_1.z.string().describe('The ID of the voice to use'),
    outputPath: zod_1.z.string().describe('The file path where the audio will be saved'),
    modelId: zod_1.z.string().optional().describe('The ID of the model to use (default: eleven_monolingual_v1)'),
    stability: zod_1.z.number().min(0).max(1).optional().describe('Voice stability (0-1, default: 0.5)'),
    similarityBoost: zod_1.z.number().min(0).max(1).optional().describe('Voice similarity boost (0-1, default: 0.5)'),
    style: zod_1.z.number().min(0).max(1).optional().describe('Voice style (0-1, for compatible models)'),
    useSpeakerBoost: zod_1.z.boolean().optional().describe('Enable speaker boost (for compatible models)'),
    outputFormat: zod_1.z.enum([
        'mp3_44100_32',
        'mp3_44100_64',
        'mp3_44100_96',
        'mp3_44100_128',
        'mp3_44100_192',
        'pcm_16000',
        'pcm_22050',
        'pcm_24000',
        'pcm_44100',
        'ulaw_8000'
    ]).optional().describe('Audio output format (default: mp3_44100_128)'),
    optimizeStreamingLatency: zod_1.z.number().min(0).max(4).optional().describe('Streaming latency optimization (0-4)')
});
exports.textToSpeechTool = {
    name: 'elevenlabs_text_to_speech',
    description: 'Convert text to speech using ElevenLabs API and save to file',
    inputSchema,
    handler: async (args) => {
        const accountState = new account_state_js_1.AccountState();
        const client = new elevenlabs_client_js_1.ElevenLabsClient(accountState);
        const options = {
            modelId: args.modelId,
            voiceSettings: {
                stability: args.stability ?? 0.5,
                similarity_boost: args.similarityBoost ?? 0.5,
                style: args.style,
                use_speaker_boost: args.useSpeakerBoost
            },
            outputFormat: args.outputFormat,
            optimizeStreamingLatency: args.optimizeStreamingLatency
        };
        try {
            const audioBuffer = await client.textToSpeech(args.text, args.voiceId, options);
            const outputDir = path_1.default.dirname(args.outputPath);
            await fs_1.promises.mkdir(outputDir, { recursive: true });
            await fs_1.promises.writeFile(args.outputPath, audioBuffer);
            return {
                content: [{
                        type: 'text',
                        text: JSON.stringify({
                            success: true,
                            outputPath: args.outputPath,
                            fileSize: audioBuffer.length,
                            format: args.outputFormat || 'mp3_44100_128',
                            textLength: args.text.length
                        }, null, 2)
                    }]
            };
        }
        catch (error) {
            return {
                content: [{
                        type: 'text',
                        text: JSON.stringify({
                            success: false,
                            error: error instanceof Error ? error.message : 'Unknown error occurred'
                        }, null, 2)
                    }]
            };
        }
    }
};
//# sourceMappingURL=text-to-speech.js.map