SenseAudio Clone Wizard

A guided experience for voice cloning. The API itself is simple — the hard part is getting a clean recording. This skill handles that.

Note: SenseAudio does not support cloning via API. The actual cloning happens on the platform at https://senseaudio.cn. This skill guides the user through the full process and handles the steps that can be automated.

Phase 1: Before Recording

When a user says they want to clone their voice, show this guide before they record anything:

录音前请注意：

• 环境：找一个安静的房间，关掉空调、风扇、电视
• 距离：嘴巴距离麦克风约 20cm，不要太近（会爆音）也不要太远（声音发虚）
• 时长：3–30 秒，朗读一段自然的文字效果最好
• 格式：MP3、WAV 或 AAC，文件不超过 50MB
• 内容：朗读一段话，语速自然，避免长时间停顿

推荐朗读内容（约 15 秒）：

"大家好，我是 [你的名字]。今天天气不错，阳光明媚。我喜欢在这样的日子里出去走走，感受大自然的美好。希望每一天都能这样轻松愉快。"

录好后把文件发给我，我来帮你检测质量。

Phase 2: Audio Quality Check

When the user uploads an audio file, run the quality check:

RESULT=$(curl -s -X POST https://api.senseaudio.cn/v1/audio/analysis \
  -H "Authorization: Bearer $SENSEAUDIO_API_KEY" \
  -F "model=sense-asr-check" \
  -F "file=@\x3CAUDIO_FILE>")

Parse the response and translate it into plain language:

Noise score interpretation:

Noise type translations:

Example diagnosis output:

你的录音检测结果：
- 时长：14.3 秒 ✓
- 噪声评分：0.52（中等）
- 检测到：嗡嗡声（可能是空调）

建议：关掉空调后重新录一段。如果条件不允许，可以继续提交，
但克隆出的音色可能带有轻微底噪。要继续还是重录？

If has_noise is false or score \x3C 0.2, skip the warning and proceed directly to Phase 3.

Phase 3: Platform Cloning

The cloning itself must be done on the SenseAudio platform — the API does not support direct cloning requests.

Guide the user:

录音质量通过！接下来去平台完成克隆：

1. 打开 https://senseaudio.cn，登录你的账号
2. 进入「音色克隆」页面
3. 点击「上传音频」，选择你刚才的录音文件
4. 系统会自动训练，通常几秒内完成
5. 训练完成后，在「我的音色」中找到新音色，复制它的 voice_id
6. 把 voice_id 发给我，我来帮你生成第一段试听

注意槽位： 每个账号有有限的克隆槽位。如果槽位已满，需要先删除旧音色才能克隆新的。

Phase 4: First Playback

Once the user provides their voice_id, synthesize a welcome message so they can hear their AI voice for the first time:

Default preview text (warm and personal):

"你好！这是我的 AI 声音。从今天起，我可以用这个声音说任何我想说的话了。"

curl -s -X POST https://api.senseaudio.cn/v1/t2a_v2 \
  -H "Authorization: Bearer $SENSEAUDIO_API_KEY" \
  -H "Content-Type: application/json" \
  -d "{
    \"model\": \"SenseAudio-TTS-1.0\",
    \"text\": \"你好！这是我的 AI 声音。从今天起，我可以用这个声音说任何我想说的话了。\",
    \"stream\": false,
    \"voice_setting\": { \"voice_id\": \"\x3CVOICE_ID>\" },
    \"audio_setting\": { \"format\": \"mp3\" }
  }" -o preview.json

jq -r '.data.audio' preview.json | xxd -r -p > my_voice_preview.mp3

After generating: tell the user the file is ready, and mention they can now use this voice_id with any SenseAudio TTS feature — including the senseaudio-tts-quick skill for quick synthesis or senseaudio-polyglot-tts for precise pronunciation control.