koishi-plugin-minimax-vits 1.2.0 → 1.2.7

package/lib/index.d.ts

@@ -1,10 +1,5 @@
 import { Context, Schema } from 'koishi';
 import { Tool } from '@langchain/core/tools';
-interface ChatLunaToolRunnable {
-    configurable: {
-        session: any;
-    };
-}
 declare module '@koishijs/plugin-console' {
     namespace Console {
         interface Services {
@@ -19,7 +14,7 @@ export declare class MinimaxVitsTool extends Tool {
     name: string;
     description: string;
     constructor(ctx: Context, config: Config, cacheManager?: AudioCacheManager | undefined);
-    _call(input: string, _runManager: any, toolConfig: ChatLunaToolRunnable): Promise<string>;
+    _call(input: string, _runManager: any, toolConfig: any): Promise<string>;
 }
 declare class MinimaxVitsService {
     private ctx;
@@ -66,11 +61,8 @@ declare class AudioCacheManager {
     private enabled;
     private maxAge;
     private maxSize;
-    private cacheMap;
-    private cleanupInterval;
     constructor(cacheDir: string, logger: any, enabled: boolean, maxAge: number, maxSize: number);
     initialize(): Promise<void>;
-    private startCleanupScheduler;
     getAudio(text: string, voice: string, format: string): Promise<Buffer | null>;
     saveAudio(buffer: Buffer, text: string, voice: string, format: string): Promise<void>;
     dispose(): void;

package/lib/index.js

@@ -1,53 +1,14 @@
-"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;
-    };
-})();
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.Config = exports.name = exports.MinimaxVitsTool = void 0;
-exports.apply = apply;
-const koishi_1 = require("koishi");
-const tools_1 = require("@langchain/core/tools");
-const fs = __importStar(require("fs"));
-const path = __importStar(require("path"));
-const crypto = __importStar(require("crypto"));
-// 引入 ChatLuna 服务类
-const chat_1 = require("koishi-plugin-chatluna/services/chat");
-// 辅助函数：模糊查询
+import { Schema, h } from 'koishi';
+import { Tool } from '@langchain/core/tools';
+import * as fs from 'fs';
+import * as path from 'path';
+import * as crypto from 'crypto';
+import { ChatLunaPlugin } from 'koishi-plugin-chatluna/services/chat';
+// --- 辅助函数 ---
 function fuzzyQuery(text, keywords) {
     const lowerText = text.toLowerCase();
     return keywords.some(keyword => lowerText.includes(keyword.toLowerCase()));
 }
-// 辅助函数：获取消息文本内容
 function getMessageContent(content) {
     if (typeof content === 'string')
         return content;
@@ -56,7 +17,6 @@ function getMessageContent(content) {
     }
     return String(content);
 }
-// 辅助函数：从长文本中提取对话内容（避免朗读旁白）
 function extractDialogueContent(text) {
     const lines = text.split('\n').map(line => line.trim()).filter(line => line.length > 0);
     let dialogueContent = '';
@@ -65,7 +25,7 @@ function extractDialogueContent(text) {
         const isDialogueLine = line.startsWith('"') ||
             line.startsWith("'") ||
             line.includes('说：') ||
-            line.match(/^[A-Za-z\u4e00-\u9fff]+[：:]/); // 简单的人名冒号匹配
+            line.match(/^[A-Za-z\u4e00-\u9fff]+[：:]/);
         const isNonDialogue = (line.includes('（') && line.includes('）')) ||
             (line.includes('(') && line.includes(')')) ||
             line.match(/^\s*[\[\{【（(]/);
@@ -85,29 +45,26 @@ function extractDialogueContent(text) {
             dialogueContent += line + '。';
         }
     }
-    if (dialogueContent.length > 0) {
+    if (dialogueContent.length > 0)
         return dialogueContent.replace(/。+/g, '。').trim();
-    }
-    // 如果没有明显对话标记且文本较短，直接朗读全文
-    if (text.length <= 150 && !text.match(/[[{【（(]/)) {
+    if (text.length <= 150 && !text.match(/[[{【（(]/))
         return text;
-    }
     return null;
 }
-// --- ChatLuna Tool 定义 ---
-class MinimaxVitsTool extends tools_1.Tool {
+// --- 辅助：构建音频消息元素 ---
+function makeAudioElement(buffer, format) {
+    const mimeType = format === 'wav' ? 'audio/wav' : 'audio/mpeg';
+    const src = `data:${mimeType};base64,${buffer.toString('base64')}`;
+    return h('audio', { src });
+}
+// --- ChatLuna Tool 类 ---
+export class MinimaxVitsTool extends Tool {
     ctx;
     config;
     cacheManager;
     name = 'minimax_tts';
-    // 提供给 LLM 的详细描述，指导其何时调用
-    description = `Use this tool to generate speech/audio from text using MiniMax TTS (Text-to-Speech).
-  Input MUST be a JSON string with the following keys:
-  - text (required): The text content to convert to speech.
-  - voice (optional): Voice ID (default is "Chinese_female_gentle").
-  - speed (optional): Speed of speech (0.5-2.0).
-  
-  Example input: "{\\"text\\": \\"Hello, how are you?\\", \\"speed\\": 1.1}"`;
+    description = `Use this tool to generate speech/audio from text using MiniMax TTS.
+  Input MUST be a JSON string: {"text": "required content", "voice": "optional_id", "speed": 1.0}`;
     constructor(ctx, config, cacheManager) {
         super();
         this.ctx = ctx;
@@ -116,14 +73,15 @@ class MinimaxVitsTool extends tools_1.Tool {
     }
     async _call(input, _runManager, toolConfig) {
         try {
-            const session = toolConfig.configurable.session;
-            const logger = this.ctx.logger('minimax-vits');
+            const session = toolConfig?.configurable?.session;
+            if (!session) {
+                throw new Error('Session not found in tool config');
+            }
             let params = {};
             try {
                 params = JSON.parse(input);
             }
             catch {
-                // 容错：如果 LLM 没传 JSON，直接当纯文本处理
                 params = { text: input };
             }
             let text = params.text || input;
@@ -131,32 +89,25 @@ class MinimaxVitsTool extends tools_1.Tool {
                 text = JSON.stringify(text);
             const voiceId = (params.voice || this.config.defaultVoice) ?? 'Chinese_female_gentle';
             const speed = params.speed ?? this.config.speed ?? 1.0;
-            const vol = params.vol ?? this.config.vol ?? 1.0;
-            const pitch = params.pitch ?? this.config.pitch ?? 0;
-            // 提取纯对话内容，优化朗读体验
             const dialogueText = extractDialogueContent(text);
-            if (!dialogueText) {
-                return `未检测到有效的对话内容，跳过语音生成。`;
-            }
-            if (this.config.debug) {
-                logger.debug(`Tool调用: voice=${voiceId}, text=${dialogueText.substring(0, 30)}...`);
-            }
-            const audioBuffer = await generateSpeech(this.ctx, { ...this.config, speed, vol, pitch }, dialogueText, voiceId, this.cacheManager);
-            if (!audioBuffer) {
-                return `TTS 生成失败，请稍后重试。`;
-            }
-            const mimeType = this.config.audioFormat === 'mp3' ? 'audio/mpeg' : 'audio/wav';
-            // 直接向用户发送音频元素
-            await session.send((0, koishi_1.h)('audio', { src: `base64://${audioBuffer.toString('base64')}`, type: mimeType }));
-            return `Successfully generated audio for: "${dialogueText}". The audio has been sent to the user.`;
+            if (!dialogueText)
+                return `未检测到有效对话内容。`;
+            const audioBuffer = await generateSpeech(this.ctx, {
+                ...this.config,
+                speed,
+                vol: params.vol,
+                pitch: params.pitch
+            }, dialogueText, voiceId, this.cacheManager);
+            if (!audioBuffer)
+                return `TTS 生成失败。`;
+            await session.send(makeAudioElement(audioBuffer, this.config.audioFormat ?? 'mp3'));
+            return `Audio generated and sent.`;
         }
         catch (e) {
-            this.ctx.logger('minimax-vits').error('Tool error:', e);
-            return `TTS Tool execution failed: ${e.message}`;
+            return `Error: ${e.message}`;
         }
     }
 }
-exports.MinimaxVitsTool = MinimaxVitsTool;
 // --- Console Service ---
 class MinimaxVitsService {
     ctx;
@@ -185,43 +136,28 @@ class MinimaxVitsService {
         }
     }
 }
-exports.name = 'minimax-vits';
-exports.Config = koishi_1.Schema.object({
-    ttsApiKey: koishi_1.Schema.string().required().description('MiniMax TTS API Key').role('secret'),
-    groupId: koishi_1.Schema.string().description('MiniMax Group ID (可选)'),
-    apiBase: koishi_1.Schema.string().default('https://api.minimax.io/v1').description('API 基础地址'),
-    defaultVoice: koishi_1.Schema.string().default('Chinese_female_gentle').description('默认语音 ID'),
-    speechModel: koishi_1.Schema.string().default('speech-01-turbo').description('TTS 模型 (推荐 speech-01-turbo, speech-01-hd)'),
-    speed: koishi_1.Schema.number().default(1.0).min(0.5).max(2.0).description('语速 (0.5-2.0)'),
-    vol: koishi_1.Schema.number().default(1.0).min(0.1).max(10.0).description('音量 (0.1-10.0)'),
-    pitch: koishi_1.Schema.number().default(0).min(-12).max(12).description('音调 (-12 到 12)'),
-    audioFormat: koishi_1.Schema.string().default('mp3').description('音频格式 (mp3, wav, flac)'),
-    sampleRate: koishi_1.Schema.number().default(32000).description('采样率'),
-    bitrate: koishi_1.Schema.number().default(128000).description('比特率'),
-    outputFormat: koishi_1.Schema.string().default('hex').description('API输出编码 (建议 hex)'),
-    languageBoost: koishi_1.Schema.string().default('auto').description('语言增强 (auto, Chinese, English)'),
-    debug: koishi_1.Schema.boolean().default(false).description('启用调试日志'),
-    voiceCloneEnabled: koishi_1.Schema.boolean().default(false).description('启用语音克隆/文件上传命令'),
-    cacheEnabled: koishi_1.Schema.boolean().default(true).description('启用本地文件缓存'),
-    cacheDir: koishi_1.Schema.string().default('./data/minimax-vits/cache').description('缓存路径'),
-    cacheMaxAge: koishi_1.Schema.number().default(3600000).description('缓存有效期(ms)'),
-    cacheMaxSize: koishi_1.Schema.number().default(104857600).description('缓存最大体积(bytes)'),
+export const name = 'minimax-vits';
+export const Config = Schema.object({
+    ttsApiKey: Schema.string().required().description('MiniMax TTS API Key').role('secret'),
+    groupId: Schema.string().description('MiniMax Group ID'),
+    apiBase: Schema.string().default('https://api.minimax.io/v1').description('API 基础地址'),
+    defaultVoice: Schema.string().default('Chinese_female_gentle').description('默认语音 ID'),
+    speechModel: Schema.string().default('speech-01-turbo').description('TTS 模型 (推荐 speech-01-turbo)'),
+    speed: Schema.number().default(1.0).min(0.5).max(2.0).description('语速'),
+    vol: Schema.number().default(1.0).description('音量'),
+    pitch: Schema.number().default(0).description('音调'),
+    audioFormat: Schema.string().default('mp3').description('音频格式 (mp3, wav)'),
+    sampleRate: Schema.number().default(32000).description('采样率'),
+    bitrate: Schema.number().default(128000).description('比特率'),
+    outputFormat: Schema.string().default('hex').description('API输出编码 (必须是 hex)'),
+    languageBoost: Schema.string().default('auto').description('语言增强'),
+    debug: Schema.boolean().default(false).description('启用调试日志'),
+    voiceCloneEnabled: Schema.boolean().default(false).description('启用语音克隆'),
+    cacheEnabled: Schema.boolean().default(true).description('启用本地文件缓存'),
+    cacheDir: Schema.string().default('./data/minimax-vits/cache').description('缓存路径'),
+    cacheMaxAge: Schema.number().default(3600000).description('缓存有效期(ms)'),
+    cacheMaxSize: Schema.number().default(104857600).description('缓存最大体积(bytes)'),
 }).description('MiniMax VITS 配置');
-// --- 音频处理辅助函数 ---
-async function decodeAudioFromHex(hexString, logger) {
-    try {
-        if (!hexString)
-            return null;
-        const buffer = Buffer.from(hexString, 'hex');
-        if (buffer.length === 0)
-            return null;
-        return buffer;
-    }
-    catch (e) {
-        logger.error('Hex 解码失败:', e.message);
-        return null;
-    }
-}
 // --- 缓存管理器 ---
 class AudioCacheManager {
     cacheDir;
@@ -229,8 +165,6 @@ class AudioCacheManager {
     enabled;
     maxAge;
     maxSize;
-    cacheMap = new Map();
-    cleanupInterval = null;
     constructor(cacheDir, logger, enabled, maxAge, maxSize) {
         this.cacheDir = cacheDir;
         this.logger = logger;
@@ -241,17 +175,8 @@ class AudioCacheManager {
     async initialize() {
         if (!this.enabled)
             return;
-        try {
-            if (!fs.existsSync(this.cacheDir))
-                fs.mkdirSync(this.cacheDir, { recursive: true });
-            this.startCleanupScheduler();
-        }
-        catch (e) {
-            this.logger.warn('缓存初始化失败', e);
-        }
-    }
-    startCleanupScheduler() {
-        this.cleanupInterval = setInterval(() => { }, 600000);
+        if (!fs.existsSync(this.cacheDir))
+            fs.mkdirSync(this.cacheDir, { recursive: true });
     }
     async getAudio(text, voice, format) {
         if (!this.enabled)
@@ -277,22 +202,17 @@ class AudioCacheManager {
             this.logger.warn('缓存写入失败', e);
         }
     }
-    dispose() {
-        if (this.cleanupInterval)
-            clearInterval(this.cleanupInterval);
-    }
+    dispose() { }
 }
-// --- 核心生成逻辑 (对接 V2 API) ---
+// --- 核心功能函数 ---
 async function generateSpeech(ctx, config, text, voice, cacheManager) {
     const logger = ctx.logger('minimax-vits');
-    const apiBase = config.apiBase ?? 'https://api.minimax.io/v1';
     const format = config.audioFormat ?? 'mp3';
-    // 1. 查缓存
     if (cacheManager) {
         const cached = await cacheManager.getAudio(text, voice, format);
         if (cached) {
             if (config.debug)
-                logger.debug('Hit cache');
+                logger.debug('命中本地缓存');
             return cached;
         }
     }
@@ -303,12 +223,11 @@ async function generateSpeech(ctx, config, text, voice, cacheManager) {
         };
         if (config.groupId)
             headers['GroupId'] = config.groupId;
-        // 2. 构造符合 T2A V2 文档的 Payload
         const payload = {
             model: config.speechModel ?? 'speech-01-turbo',
             text: text,
-            stream: false, // 强制关闭流式以简化处理
-            output_format: config.outputFormat ?? 'hex', // 推荐使用 hex
+            stream: false,
+            output_format: 'hex',
             voice_setting: {
                 voice_id: voice,
                 speed: config.speed ?? 1.0,
@@ -325,107 +244,175 @@ async function generateSpeech(ctx, config, text, voice, cacheManager) {
         if (config.languageBoost && config.languageBoost !== 'auto') {
             payload.language_boost = config.languageBoost;
         }
-        if (config.debug) {
-            logger.debug(`POST ${apiBase}/t2a_v2`);
-            logger.debug(`Payload: ${JSON.stringify(payload)}`);
-        }
-        // 3. 发起请求
-        const response = await ctx.http.post(`${apiBase}/t2a_v2`, payload, { headers, timeout: 60000 });
-        // 4. 检查响应状态
+        if (config.debug)
+            logger.debug(`调用 API: ${config.apiBase}/t2a_v2`);
+        const response = await ctx.http.post(`${config.apiBase}/t2a_v2`, payload, { headers, timeout: 60000 });
         if (response?.base_resp && response.base_resp.status_code !== 0) {
-            logger.error(`API Error: [${response.base_resp.status_code}] ${response.base_resp.status_msg}`);
+            logger.error(`API Error: ${response.base_resp.status_msg}`);
             return null;
         }
-        // 5. 解析音频数据 (优先 data.audio，兼容部分 SDK 的扁平化处理)
         const audioHex = response?.data?.audio || response?.audio;
         if (!audioHex) {
-            logger.error('API 响应中未找到音频数据 (response.data.audio)');
-            if (config.debug)
-                logger.debug('Response:', JSON.stringify(response));
+            logger.error('API 返回数据中未找到 audio 字段');
             return null;
         }
-        // 6. 解码 Hex
-        const audioBuffer = await decodeAudioFromHex(audioHex, logger);
-        // 7. 写入缓存
-        if (audioBuffer && cacheManager) {
+        const audioBuffer = Buffer.from(audioHex, 'hex');
+        if (audioBuffer.length === 0)
+            return null;
+        if (cacheManager) {
             await cacheManager.saveAudio(audioBuffer, text, voice, format);
         }
         return audioBuffer;
     }
     catch (error) {
-        logger.error('TTS 请求失败:', error);
-        if (error.response?.data) {
-            logger.error('API Error Detail:', JSON.stringify(error.response.data));
-        }
+        logger.error('TTS 调用失败:', error);
         return null;
     }
 }
-// --- 文件上传逻辑 ---
+// 修正：返回值类型改为 string | undefined，匹配调用处的类型
 async function uploadFile(ctx, config, filePath, purpose) {
-    const headers = { 'Authorization': `Bearer ${config.ttsApiKey}` };
-    if (config.groupId)
-        headers['GroupId'] = config.groupId;
-    const formData = new FormData();
-    formData.append('file', await ctx.http.file(filePath));
-    formData.append('purpose', purpose);
-    const res = await ctx.http.post(`${config.apiBase}/files/upload`, formData, { headers });
-    return res.file?.file_id;
+    const logger = ctx.logger('minimax-vits');
+    try {
+        const headers = { 'Authorization': `Bearer ${config.ttsApiKey}` };
+        if (config.groupId)
+            headers['GroupId'] = config.groupId;
+        const fileRes = await ctx.http.file(filePath);
+        // 如果没有 Blob 类型（Node 低版本），需要 polyfill 或者忽略类型报错
+        // 此处假设环境支持，使用 new Blob 包装 buffer
+        const blob = new Blob([fileRes.data], { type: fileRes.mime });
+        const formData = new FormData();
+        formData.append('file', blob, fileRes.filename || 'upload.mp3');
+        formData.append('purpose', purpose);
+        const response = await ctx.http.post(`${config.apiBase}/files/upload`, formData, { headers });
+        return response.file?.file_id || undefined;
+    }
+    catch (error) {
+        logger.error(`文件上传失败:`, error);
+        return undefined;
+    }
 }
-// --- 语音克隆逻辑 ---
-async function cloneVoice(ctx, config, fileId, voiceId, text) {
-    // 注意：MiniMax 克隆接口参数可能会变动，这里保持基础实现
-    const headers = { 'Authorization': `Bearer ${config.ttsApiKey}`, 'Content-Type': 'application/json' };
-    if (config.groupId)
-        headers['GroupId'] = config.groupId;
-    const payload = {
-        file_id: fileId,
-        voice_id: voiceId,
-        model: config.speechModel,
-        text: text,
-        audio_format: config.audioFormat ?? 'mp3'
-    };
-    const res = await ctx.http.post(`${config.apiBase}/voice_clone`, payload, { headers, responseType: 'arraybuffer' });
-    return Buffer.from(res);
+// 语音克隆逻辑
+async function cloneVoice(ctx, config, fileId, voiceId, promptAudioFileId, promptText, text) {
+    const logger = ctx.logger('minimax-vits');
+    try {
+        const payload = {
+            file_id: fileId,
+            voice_id: voiceId,
+            model: config.speechModel ?? 'speech-01-turbo',
+            audio_format: config.audioFormat ?? 'mp3',
+        };
+        if (text)
+            payload.text = text;
+        if (promptAudioFileId && promptText) {
+            payload.clone_prompt = { prompt_audio: promptAudioFileId, prompt_text: promptText };
+        }
+        const headers = {
+            'Authorization': `Bearer ${config.ttsApiKey}`,
+            'Content-Type': 'application/json',
+        };
+        if (config.groupId)
+            headers['GroupId'] = config.groupId;
+        const response = await ctx.http.post(`${config.apiBase}/voice_clone`, payload, { headers, responseType: 'arraybuffer' });
+        return Buffer.from(response);
+    }
+    catch (error) {
+        logger.error('语音克隆失败:', error);
+        return null;
+    }
 }
 // --- 插件入口 ---
-function apply(ctx, config) {
+export function apply(ctx, config) {
     const logger = ctx.logger('minimax-vits');
-    // 1. 初始化 ChatLuna 插件服务 (关键：参数 false 表示不作为模型适配器，仅作为工具集)
-    const chatLunaPlugin = new chat_1.ChatLunaPlugin(ctx, config, 'minimax-vits', false);
+    // 修正：这里使用 config as any 规避类型检查，因为 ChatLunaPlugin 需要的某些配置字段（如 proxy）我们没有定义
+    const chatLunaPlugin = new ChatLunaPlugin(ctx, config, 'minimax-vits', false);
     const cacheManager = config.cacheEnabled
         ? new AudioCacheManager(config.cacheDir ?? './data/minimax-vits/cache', logger, true, config.cacheMaxAge ?? 3600000, config.cacheMaxSize ?? 104857600)
         : undefined;
     ctx.on('ready', async () => {
         await cacheManager?.initialize();
-        // 2. 注册控制台服务
         if (ctx.console) {
             ctx.console.services['minimax-vits'] = new MinimaxVitsService(ctx, config);
         }
-        // 3. 注册 ChatLuna 工具
         try {
             chatLunaPlugin.registerTool('minimax_tts', {
                 selector: (history) => history.some((item) => fuzzyQuery(getMessageContent(item.content), ['语音', '朗读', 'tts', 'speak', 'say', 'voice'])),
                 createTool: () => new MinimaxVitsTool(ctx, config, cacheManager),
                 authorization: () => true
             });
-            logger.info('ChatLuna Tool "minimax_tts" 已注册');
+            logger.info('ChatLuna Tool 已注册');
         }
         catch (e) {
-            logger.warn('ChatLuna Tool 注册失败 (可能是 chatluna 插件未安装):', e.message);
+            logger.warn('ChatLuna Tool 注册失败', e.message);
         }
     });
     ctx.on('dispose', () => cacheManager?.dispose());
-    // 注册常规指令
+    // --- 指令注册区 ---
     ctx.command('minivits.test <text:text>', '测试 TTS')
         .option('voice', '-v <voice>')
+        .option('speed', '-s <speed>', { type: 'number' })
         .action(async ({ session, options }, text) => {
         if (!text)
             return '请输入文本';
         await session?.send('生成中...');
-        const buffer = await generateSpeech(ctx, config, text, options?.voice || config.defaultVoice || 'Chinese_female_gentle', cacheManager);
+        const buffer = await generateSpeech(ctx, {
+            ...config,
+            speed: options?.speed ?? config.speed
+        }, text, options?.voice || config.defaultVoice || 'Chinese_female_gentle', cacheManager);
         if (!buffer)
             return '失败';
-        return (0, koishi_1.h)('audio', { src: `base64://${buffer.toString('base64')}`, type: config.audioFormat === 'mp3' ? 'audio/mpeg' : 'audio/wav' });
+        return makeAudioElement(buffer, config.audioFormat ?? 'mp3');
+    });
+    ctx.command('minivits.debug', '查看插件配置').action(() => {
+        return `API Base: ${config.apiBase}\nModel: ${config.speechModel}\nFormat: ${config.audioFormat}\nDebug: ${config.debug}`;
     });
-    // 克隆指令略 (保持原样即可)
+    if (config.voiceCloneEnabled) {
+        ctx.command('minivits.clone.upload <filePath> <purpose>', '上传文件')
+            .action(async ({ session }, filePath, purpose) => {
+            if (!session || !filePath || !purpose)
+                return '缺少参数';
+            if (purpose !== 'voice_clone' && purpose !== 'prompt_audio')
+                return '用途错误';
+            await session.send('上传中...');
+            const fileId = await uploadFile(ctx, config, filePath, purpose);
+            return fileId ? `上传成功: ${fileId}` : '上传失败';
+        });
+        ctx.command('minivits.clone.create <fileId> <voiceId> [text:text]', '创建语音克隆')
+            .option('promptAudio', '-p <id>')
+            .option('promptText', '-t <text>')
+            .action(async ({ session, options }, fileId, voiceId, text) => {
+            if (!session || !fileId || !voiceId)
+                return '缺少参数';
+            await session.send('克隆中...');
+            const audioBuffer = await cloneVoice(ctx, config, fileId, voiceId, options?.promptAudio, options?.promptText, text);
+            if (!audioBuffer)
+                return '克隆失败';
+            if (text) {
+                return makeAudioElement(audioBuffer, config.audioFormat ?? 'mp3');
+            }
+            return '克隆操作请求已发送';
+        });
+        ctx.command('minivits.clone.full <sourceFile> <voiceId> <text:text>', '完整克隆流程')
+            .option('promptFile', '-p <file>')
+            .option('promptText', '-t <text>')
+            .action(async ({ session, options }, sourceFile, voiceId, text) => {
+            if (!session || !sourceFile || !voiceId || !text)
+                return '缺少参数';
+            await session.send('1. 上传源文件...');
+            const sourceFileId = await uploadFile(ctx, config, sourceFile, 'voice_clone');
+            if (!sourceFileId)
+                return '源文件上传失败';
+            let promptAudioFileId;
+            if (options?.promptFile) {
+                await session.send('2. 上传提示音频...');
+                promptAudioFileId = await uploadFile(ctx, config, options.promptFile, 'prompt_audio');
+                if (!promptAudioFileId)
+                    return '提示音频上传失败';
+            }
+            await session.send('3. 生成克隆语音...');
+            const audioBuffer = await cloneVoice(ctx, config, sourceFileId, voiceId, promptAudioFileId, options?.promptText, text);
+            if (!audioBuffer)
+                return '语音克隆失败';
+            return makeAudioElement(audioBuffer, config.audioFormat ?? 'mp3');
+        });
+    }
 }

package/package.json

@@ -1,7 +1,8 @@
 {
   "name": "koishi-plugin-minimax-vits",
   "description": "使用 minimax 国际版生成语音，适配 chatluna",
-  "version": "1.2.0",
+  "version": "1.2.7",
+  "type": "module",
   "main": "lib/index.js",
   "typings": "lib/index.d.ts",
   "files": [
@@ -23,7 +24,10 @@
     "koishi": "^4.18.10"
   },
   "inject": {
-    "optional": ["console", "chatluna"]
+    "optional": [
+      "console",
+      "chatluna"
+    ]
   },
   "devDependencies": {
     "@types/node": "^20.0.0",