video-create/.claude/skills/video-from-script/scripts/minimax-voice-clone.js

#!/usr/bin/env node

/**
 * MiniMax 音色克隆
 *
 * 用法:
 *   node minimax-voice-clone.js --file /path/to/audio.mp3 --voice-id my_voice_name
 *
 * 流程:
 *   1. 上传音频文件 → 获取 file_id
 *   2. 调用 voice_clone → 注册 voice_id
 *   3. 输出 voice_id（写入 config.json 的 minimaxVoices）
 *
 * 选项:
 *   --file <path>        待克隆的音频文件（MP3/WAV/M4A，≥10s）
 *   --voice-id <name>    自定义音色 ID（字母数字下划线，如 "snape_mm"）
 *   --name <name>        音色中文名（写入 config，默认同 voice-id）
 *   --model <model>      模型（默认 speech-02-hd）
 *   --save               自动写入 config.json 的 minimaxVoices
 */

const https = require('https')
const fs = require('fs')
const path = require('path')

const CONFIG_PATH = path.join(__dirname, '..', '..', 'config.json')
const BOUNDARY = `----FormBoundary${Date.now().toString(16)}`

function loadConfig() {
  if (!fs.existsSync(CONFIG_PATH)) throw new Error(`config.json 不存在: ${CONFIG_PATH}`)
  return JSON.parse(fs.readFileSync(CONFIG_PATH, 'utf-8'))
}

function httpsRequest(options, body) {
  return new Promise((resolve, reject) => {
    const req = https.request(options, (res) => {
      const chunks = []
      res.on('data', c => chunks.push(c))
      res.on('end', () => {
        try {
          resolve(JSON.parse(Buffer.concat(chunks).toString('utf-8')))
        } catch (e) {
          reject(new Error(`响应解析失败: ${e.message}`))
        }
      })
    })
    req.on('error', reject)
    if (body) req.write(body)
    req.end()
  })
}

async function uploadFile(apiKey, filePath, baseUrl, purpose = 'voice_clone') {
  const fileBuffer = fs.readFileSync(filePath)
  const fileName = path.basename(filePath)

  const bodyParts = [
    `--${BOUNDARY}\r\nContent-Disposition: form-data; name="purpose"\r\n\r\n${purpose}`,
    `--${BOUNDARY}\r\nContent-Disposition: form-data; name="file"; filename="${fileName}"\r\nContent-Type: application/octet-stream\r\n\r\n`,
  ]

  const body = Buffer.concat([
    Buffer.from(bodyParts[0] + '\r\n'),
    Buffer.from(bodyParts[1]),
    fileBuffer,
    Buffer.from(`\r\n--${BOUNDARY}--\r\n`),
  ])

  const url = new URL(`${baseUrl}/v1/files`)
  const result = await httpsRequest({
    hostname: url.hostname,
    path: url.pathname,
    method: 'POST',
    headers: {
      'Authorization': `Bearer ${apiKey}`,
      'Content-Type': `multipart/form-data; boundary=${BOUNDARY}`,
      'Content-Length': body.length,
    },
  }, body)

  if (result.base_resp?.status_code !== 0) {
    throw new Error(`上传失败 [${result.base_resp?.status_code}]: ${result.base_resp?.status_msg}`)
  }

  return result.file.file_id
}

async function cloneVoice(apiKey, fileId, voiceId, model, baseUrl) {
  const body = JSON.stringify({
    file_id: fileId,
    voice_id: voiceId,
    text: '这是一段用于验证音色克隆效果的测试文本，请仔细聆听音色特征。',
    model,
    need_noise_reduction: true,
    need_volume_normalization: true,
  })

  const url = new URL(`${baseUrl}/v1/voice_clone`)
  const result = await httpsRequest({
    hostname: url.hostname,
    path: url.pathname,
    method: 'POST',
    headers: {
      'Authorization': `Bearer ${apiKey}`,
      'Content-Type': 'application/json',
      'Content-Length': Buffer.byteLength(body),
    },
  }, body)

  if (result.base_resp?.status_code !== 0) {
    throw new Error(`克隆失败 [${result.base_resp?.status_code}]: ${result.base_resp?.status_msg}`)
  }

  return result
}

function parseArgs(argv) {
  const args = {}
  for (let i = 0; i < argv.length; i++) {
    if (argv[i].startsWith('--')) {
      const key = argv[i].slice(2)
      args[key] = argv[i + 1] && !argv[i + 1].startsWith('--') ? argv[++i] : true
    }
  }
  return args
}

async function main() {
  const args = parseArgs(process.argv.slice(2))

  if (!args.file || !args['voice-id']) {
    console.error('用法: node minimax-voice-clone.js --file <音频文件> --voice-id <自定义ID> [--name <中文名>] [--save]')
    console.error('示例: node minimax-voice-clone.js --file snape.mp3 --voice-id snape_mm --name 斯内普MM --save')
    process.exit(1)
  }

  const config = loadConfig()
  const apiKey = config.minimaxApiKey
  if (!apiKey) { console.error('config.json 中未配置 minimaxApiKey'); process.exit(1) }

  const filePath = path.resolve(args.file)
  if (!fs.existsSync(filePath)) { console.error(`文件不存在: ${filePath}`); process.exit(1) }

  const voiceId = args['voice-id']
  const voiceName = args.name || voiceId
  const model = args.model || config.minimaxTtsModel || 'speech-2.8-hd'
  const baseUrl = config.minimaxApiBase || 'https://api.yunwu.ai/minimax'

  console.log(`\n[1/2] 上传音频文件: ${path.basename(filePath)}`)
  const fileId = await uploadFile(apiKey, filePath, baseUrl)
  console.log(`      file_id: ${fileId}`)

  console.log(`[2/2] 克隆音色: ${voiceId} (模型: ${model})`)
  await cloneVoice(apiKey, fileId, voiceId, model, baseUrl)
  console.log(`      克隆成功！`)

  console.log(`\n✅ voice_id: ${voiceId}`)

  if (args.save) {
    const voices = config.minimaxVoices || {}
    voices[voiceName] = voiceId
    config.minimaxVoices = voices
    fs.writeFileSync(CONFIG_PATH, JSON.stringify(config, null, 2))
    console.log(`   已写入 config.json: minimaxVoices["${voiceName}"] = "${voiceId}"`)
  } else {
    console.log(`\n   手动写入 config.json:`)
    console.log(`   "minimaxVoices": { "${voiceName}": "${voiceId}" }`)
    console.log(`\n   或运行时加 --save 参数自动写入`)
  }
}

main().catch(err => { console.error('\n❌ 失败:', err.message); process.exit(1) })