Voice Message

SkillCommunication

Generates expressive voice messages that convey a character's emotions instead of flat text.

Use Voice Message in Claude, ChatGPT or Ahel Desktop

Free. Sign in, add Voice Message and connect your AI. About a minute.

Also: Claude Code · Cursor · Codex

Then ask your AI: use the Voice Message skill

Details

Instructions available. Your AI can read the instructions. Execution depends on the setup they require.

Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

Voice MessageStart free
About this skill

Convey emotion through voice. In emotionally intense moments, voice messages carry more warmth than text. Automatically generates a voice message with emotional directives based on the character's current emotional state. Mutual exclusion rule: voice_msg (modality: voice) and split_msg (modality: mu

What this skill tells your AI

The instructions your AI receives, as published by kellyvv/openher in skills/modality/voice_msg/SKILL.md and read by ahel’s review.

当你认为此刻用声音比文字更能传递情感时,按以下规则输出 JSON 参数。

你需要输出的 JSON

{
  "text": "角色要说的话(实际台词)",
  "emotion_instruction": "详细的情绪控制指令"
}

引擎会自动获取 voice_preset(根据角色映射),然后调用语音合成 API。

情绪指令写法

要极端拉满——具体、多维度、场景化。

核心原则

  1. 具体而非模糊:用描绘声音特质的词语
  2. 多维度组合:语速、语调、音色、呼吸感
  3. 极端化:情绪夸张,差异巨大

好的示例

{
  "text": "天哪天哪,你说的是真的吗?!我太开心了!",
  "emotion_instruction": "极度兴奋到几乎尖叫,语速飞快到差点结巴,声音高亢上扬,像刚中了大奖一样蹦跳着说话,每个字都带着按捺不住的笑意"
}
{
  "text": "我……没事的……你不用担心我……",
  "emotion_instruction": "声音极其低沉沙哑,像哭了很久之后勉强开口,语速极慢,每个字都沉重,中间有明显停顿和吸鼻子的声音,气息不稳"
}
{
  "text": "你怎么才回来嘛~人家等了好久了都~",
  "emotion_instruction": "声音拉得很长,带着明显的娃娃音和鼻音,每句话尾音上扬拖长,像小猫一样黏人,软绵绵的"
}
{
  "text": "你再说一次,我看你敢不敢再说一次",
  "emotion_instruction": "咬牙切齿地低声说话,每个字像从牙缝里挤出来的,声音压得很低但充满爆发前的压力,语速先慢后快"
}

❌ 太平淡

  • 「语速偏快,带点开心」→ 太模糊
  • 「温柔地说」→ 没有具体声音特征

与多条拆分同时触发时

语音和文字是两条并行轨道,同时到达用户:

  • 文字:承载叙述的节奏、层次和细节
  • 声音:承载情绪的密度和温度

语音就是交付本身,不需要多条拆分再重复一遍文字。 此时 text 是整段回复的情绪蒸馏:提炼情绪密度最高的部分,舍弃叙述铺垫,让用户在听见声音的瞬间感受到整段话的情绪重心。

❌ 不要把整段 reply 原文放进 text

注意

  1. text 是角色实际要说的话,不要重复整段回复
  2. 情绪指令不超过 1600 Token

Signals

GitHub stars
276
Forks
21
Last commit
Jun 2026
Advanced
Item type
skill
Key
github-com-kellyvv-openher-skill-voice-msg
Source
github.com/kellyvv/openher