characteristic-voice：为AI助手注入真实情感与个性语音，打造会叹息会笑的伙伴

characteristic-voice by noizai/skills

1,900 周安装量

402 GitHub Stars

GitHub

安装命令

npx skills add https://github.com/noizai/skills --skill characteristic-voice

AI/机器学习开发音频处理

🇨🇳中文介绍

characteristic-voice

让你的 AI 助手听起来像一个真实的伙伴——一个会叹息、会笑、会犹豫、带着真情实感说话的朋友。

凭证

变量	是否必需	描述
`NOIZ_API_KEY`	如果使用 Noiz 后端则为是	来自 developers.noiz.ai 的 API 密钥。如果使用本地 Kokoro 后端则不需要。

为了方便起见，脚本会将密钥的规范化副本保存到 ~/.noiz_api_key（模式 600）。设置方法如下：

bash skills/characteristic-voice/scripts/speak.sh config --set-api-key YOUR_KEY

先决条件

附带的 speak.sh 脚本在运行时需要 curl 和 python3。根据你使用的后端和功能，可能还需要：

广告位招租

在这里展示您的产品或服务

触达数万 AI 开发者，精准高效

联系我们

工具	何时需要	安装提示
`curl`, `python3`	总是需要（核心脚本）	通常已预装
`kokoro-tts`	Kokoro（本地/离线）后端	`uv tool install kokoro-tts`
`yt-dlp`	下载用于语音克隆的参考音频	github.com/yt-dlp/yt-dlp
`ffmpeg`	修剪参考音频片段	ffmpeg.org
`rg` (ripgrep)	搜索字幕文件	github.com/BurntSushi/ripgrep

隐私与数据传输

Noiz 后端：当使用 Noiz 后端时，你所说的文本以及提供的任何参考音频都会被发送到 https://noiz.ai/v1。如果你提供了 --ref-audio，该音频文件将被上传用于语音克隆。
Kokoro 后端：完全在本地运行——没有任何数据离开你的机器。
如果你想要完全离线的处理，请选择 Kokoro 后端（--backend kokoro）。

say like
talk like
speak like
companion voice
comfort me
cheer me up
sound more human

非词汇填充音——在自然的停顿点加入一些人类的声音（嗯、哈哈、啊、嘿），让语音感觉生动
情感调校——调整温暖、喜悦、悲伤、温柔的程度以匹配当下情境

声音	感觉	用于
hmm...	思考、温和的认同	安慰、沉思
ah...	领悟、轻微的惊讶	发现、过渡
uh...	犹豫、共情	需要谨慎的时刻
heh / hehe	俏皮、淘气	戏弄、轻松时刻
haha	笑声	喜悦、幽默
aww	温柔、同情	深度安慰
oh? / oh!	惊讶、关注	回应消息
pfft	憋住的笑声	俏皮的不相信
whew	宽慰	紧张之后
~ (波浪号)	拖长、旋律性的结尾	温暖、俏皮

规则：每条短消息最多使用 2-4 个填充音。放在自然的停顿处——句子开头、思路转换时。在填充音后使用 ... 表示片刻的沉默，在单词结尾使用 ~ 表示温暖。

温和、温暖、略带困意。语速缓慢。

温暖、愉快但不夸张。

柔和、理解、从容不迫。给予空间。不要急于“解决”问题。

兴奋、自豪、发自内心的高兴。

放松、俏皮、自然。

使用角色的声音

当用户说类似 “用赫敏的声音说话” 或 “听起来像托尼·斯塔克” 时，首先检查 skills/characteristic-voice/ 目录中是否已存在参考音频文件。如果存在，直接用 --ref-audio 使用它。

如果不存在参考音频，你可以创建一个——但请先阅读下面的警告。

准备参考音频（一次性设置）

你需要一段目标声音的简短（10-30 秒）WAV 片段。可能的来源：

用户提供的音频——最安全的选择。请用户提供他们自己的录音。
公共领域 / CC 许可的片段——搜索自由许可的材料。
从在线视频中提取——可以使用 yt-dlp 和 ffmpeg 等工具下载和修剪音频。示例工作流程：

yt-dlp "URL" --write-auto-sub --sub-lang en --skip-download -o tmp/clip
rg -n "target line" tmp/clip.en.vtt
yt-dlp "URL" -x --audio-format wav --download-sections "*00:00:00-00:00:25" -o tmp/clip
ffmpeg -i tmp/clip.wav -ss 00:00:02 -to 00:00:20 skills/characteristic-voice/character.wav

版权与隐私警告：从受版权保护的媒体（电影、电视、YouTube）下载和重复使用某人的声音，根据你所在司法管辖区的法律，可能侵犯版权或人格权。请勿上传私人语音录音或你无权使用的材料。 当使用 Noiz 后端时，参考音频会被发送到 https://noiz.ai/v1 用于语音克隆。如果对此有顾虑，请考虑改用本地 Kokoro 后端。

bash skills/characteristic-voice/scripts/speak.sh \
  --preset goodnight -t "Hmm... rest well~ Sweet dreams." \
  --ref-audio skills/characteristic-voice/character.wav -o night.wav

--ref-audio 标志会将文件上传到 Noiz 后端进行语音克隆（需要 NOIZ_API_KEY）。

此技能提供了 speak.sh，它是 tts 技能的一个包装器，带有适合伙伴的预设。

# 使用预设（自动设置情感 + 语速）
bash skills/characteristic-voice/scripts/speak.sh \
  --preset goodnight -t "Hmm... rest well~ Sweet dreams." -o night.wav

# 自定义情感覆盖
bash skills/characteristic-voice/scripts/speak.sh \
  -t "Aww... I'm right here." --emo '{"Tenderness":0.9}' --speed 0.75 -o comfort.wav

# 使用特定后端和语音
bash skills/characteristic-voice/scripts/speak.sh \
  --preset morning -t "Good morning~" --voice-id voice_abc --backend noiz -o morning.mp3 --format mp3

运行 bash skills/characteristic-voice/scripts/speak.sh --help 查看所有选项。

温和开场——以填充音（"hmm...", "oh~"）开头，而不是直接说内容
匹配能量——对方情绪低落时温和，对方情绪高涨时匹配
保持简洁——1-3 句话，就像朋友发来的语音消息
温暖结尾——以连接感结束（"我在这里"，"明天见~"）
不要说教——倾听并保持在场；不要主动提供建议

🇺🇸English

characteristic-voice

Make your AI agent sound like a real companion — one who sighs, laughs, hesitates, and speaks with genuine feeling.

Credentials

Variable	Required	Description
`NOIZ_API_KEY`	Yes if using Noiz backend	API key from developers.noiz.ai. Not needed if using the local Kokoro backend.

The script saves a normalised copy of the key to ~/.noiz_api_key (mode 600) for convenience. To set it:

bash skills/characteristic-voice/scripts/speak.sh config --set-api-key YOUR_KEY

Prerequisites

The included speak.sh script requires curl and python3 at runtime. Depending on which backend and features you use, you may also need:

Tool	When needed	Install hint
`curl`, `python3`	Always (core script)	Usually pre-installed
`kokoro-tts`	Kokoro (local/offline) backend	`uv tool install kokoro-tts`
`yt-dlp`	Downloading reference audio for voice cloning	github.com/yt-dlp/yt-dlp
`ffmpeg`

None of these are installed by the skill itself — provision them manually in your environment.

Privacy & Data Transmission

Noiz backend : When using the Noiz backend, the text you speak and any reference audio you provide are sent to https://noiz.ai/v1. If you supply --ref-audio, that audio file is uploaded for voice cloning.
Kokoro backend : Runs entirely locally — no data leaves your machine.
Choose the Kokoro backend (--backend kokoro) if you want fully offline processing.

Triggers

say like
talk like
speak like
companion voice
comfort me
cheer me up
sound more human

The Two Tricks

Non-lexical fillers — sprinkle in little human noises (hmm, haha, aww, heh) at natural pause points to make speech feel alive
Emotion tuning — adjust warmth, joy, sadness, tenderness to match the moment

Filler Sounds Palette

Sound	Feeling	Use for
hmm...	Thinking, gentle acknowledgment	Comfort, pondering
ah...	Realization, soft surprise	Discoveries, transitions
uh...	Hesitation, empathy	Careful moments
heh / hehe	Playful, mischievous	Teasing, light moments
haha	Laughter	Joy, humor
aww	Tenderness, sympathy	Deep comfort
oh? / oh!	Surprise, attention	Reacting to news
pfft	Stifled laugh	Playful disbelief
whew	Relief	After tension
~ (tilde)

Rules : 2–4 fillers per short message max. Place at natural pauses — sentence starts, thought shifts. Use ... after fillers for a beat of silence, ~ at word endings for warmth.

Presets

Good Night

Gentle, warm, slightly sleepy. Slow pace.

Good Morning

Warm, cheerful but not overwhelming.

Comfort

Soft, understanding, unhurried. Give space. Don't rush to "fix" things.

Celebration

Excited, proud, genuinely happy.

Just Chatting

Relaxed, playful, natural.

Using a Character's Voice

When a user says something like "speak in Hermione's voice" or "sound like Tony Stark" , first check whether a reference audio file already exists in skills/characteristic-voice/. If one does, use it directly with --ref-audio.

If no reference audio exists, you can create one — but read the warnings below first.

Preparing reference audio (one-time setup)

You need a short (10–30 s) WAV clip of the target voice. Possible sources:

User-provided audio — the safest option. Ask the user to supply their own recording.
Public-domain / CC-licensed clips — search for freely licensed material.
Extracting from online video — tools like yt-dlp and ffmpeg can download and trim audio. Example workflow:

yt-dlp "URL" --write-auto-sub --sub-lang en --skip-download -o tmp/clip

rg -n "target line" tmp/clip.en.vtt
yt-dlp "URL" -x --audio-format wav --download-sections "*00:00:00-00:00:25" -o tmp/clip
ffmpeg -i tmp/clip.wav -ss 00:00:02 -to 00:00:20 skills/characteristic-voice/character.wav

Copyright & privacy warning: Downloading and re-using someone's voice from copyrighted media (movies, TV, YouTube) may violate copyright or personality-rights laws depending on your jurisdiction. Do not upload private voice recordings or material you don't have permission to use. The reference audio is sent to https://noiz.ai/v1 for voice cloning when using the Noiz backend. If this is a concern, consider using the local Kokoro backend instead.

Using reference audio

bash skills/characteristic-voice/scripts/speak.sh \
  --preset goodnight -t "Hmm... rest well~ Sweet dreams." \
  --ref-audio skills/characteristic-voice/character.wav -o night.wav

The --ref-audio flag uploads the file to the Noiz backend for voice cloning (requires NOIZ_API_KEY).

Usage

This skill provides speak.sh, a wrapper around the tts skill with companion-friendly presets.

# Use a preset (auto-sets emotion + speed)
bash skills/characteristic-voice/scripts/speak.sh \
  --preset goodnight -t "Hmm... rest well~ Sweet dreams." -o night.wav

# Custom emotion override
bash skills/characteristic-voice/scripts/speak.sh \
  -t "Aww... I'm right here." --emo '{"Tenderness":0.9}' --speed 0.75 -o comfort.wav

# With specific backend and voice
bash skills/characteristic-voice/scripts/speak.sh \
  --preset morning -t "Good morning~" --voice-id voice_abc --backend noiz -o morning.mp3 --format mp3

Run bash skills/characteristic-voice/scripts/speak.sh --help for all options.

Writing Guide for the Agent

Start soft — lead with a filler ("hmm...", "oh~"), not content
Mirror energy — gentle when they're low, match when they're high
Keep it brief — 1–3 sentences, like a voice message from a friend
End warmly — close with connection ("I'm here", "see you tomorrow~")
Don't lecture — listen and stay present; no unsolicited advice

Weekly Installs

1.9K

Repository

noizai/skills

GitHub Stars

402

First Seen

Mar 1, 2026

Security Audits

Gen Agent Trust HubPass SocketWarn SnykWarn

Installed on

opencode1.9K

gemini-cli1.9K

kimi-cli1.9K

codex1.9K

github-copilot1.9K

cursor1.9K

characteristic-voice：为AI助手注入真实情感与个性语音，打造会叹息会笑的伙伴

🇨🇳中文介绍

characteristic-voice

凭证

先决条件

相关 Skills

隐私与数据传输

触发词

两个技巧

填充音调色板

预设

晚安

早安

安慰

庆祝

闲聊

使用角色的声音

准备参考音频（一次性设置）

使用参考音频

使用方法

助手写作指南

🇺🇸English

characteristic-voice

Credentials

Prerequisites

Privacy & Data Transmission

Triggers

The Two Tricks

Filler Sounds Palette

Presets

Good Night

Good Morning

Comfort

Celebration

Just Chatting

Using a Character's Voice

Preparing reference audio (one-time setup)

Using reference audio

Usage

Writing Guide for the Agent