voice-extractor
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseVoice Extractor
写作风格提取器
AI-generated content all sounds the same. The fix isn't better prompts — it's teaching the AI how you actually communicate.
This skill extracts your communication DNA from writing samples and produces a Voice Guide: documented, tested, and ready to use.
AI生成的内容听起来都千篇一律，解决办法不是优化提示词——而是教会AI你真实的表达方式。
本工具可以从写作样本中提取你的表达特征，生成一份经测试、可直接使用的《风格指南》。
Mode
模式
Detect from context or ask: "Quick voice snapshot, full Voice Guide, or full guide with examples?"
| Mode | What you get | Best for |
|---|---|---|
| Top 5 voice characteristics + 3 do/don't rules | Fast style reference, single piece |
| Full Voice Guide: tone, vocabulary, rhythm, structure | AI training, ghostwriting, brand documentation |
| Full Voice Guide + 10 sample rewrites + writing rules checklist + AI training examples | Onboarding writers, building a brand voice system |
Default: — use if they just need a fast reference. Use if they're onboarding a ghostwriter or building a content team.
standardquickdeep根据上下文判断，或直接询问用户：「需要快速风格快照、完整风格指南，还是带示例的完整版指南？」
| 模式 | 输出内容 | 适用场景 |
|---|---|---|
| 前5大风格特征 + 3条可做/不可做规则 | 快速风格参考、单篇内容写作 |
| 完整风格指南：语气、词汇、节奏、结构 | AI训练、代笔写作、品牌文档记录 |
| 完整风格指南 + 10个改写示例 + 写作规则清单 + AI训练示例 | 写手上岗培训、搭建品牌风格体系 |
默认： —— 若用户仅需要快速参考则使用模式，若用户要培训代笔者或搭建内容团队则使用模式。
standardquickdeepContext Loading Gates
上下文加载门槛
Before extracting, collect:
- Writing samples — minimum 3 samples OR 500 total words (see priority list below)
- Purpose of voice guide — AI training? Ghostwriter onboarding? Team alignment?
- Confidence zones — Any topics where they want to sound more/less authoritative?
- Known anti-patterns — Any words or phrases they already know they want to avoid?
Sample priority (most → least authentic):
- Casual Slack or email (raw, unedited voice)
- Podcast or call transcript
- LinkedIn posts or articles
- Website copy (often edited, less authentic)
Minimum sample gate: If samples total under 500 words, stop:
"These samples are too short to extract reliable patterns. Please add 2-3 more — emails, Slack messages, or transcripts work best. The messier and more casual, the better."
Do not attempt full extraction from under 500 words. Offer quick mode instead.
提取前请收集以下信息：
- 写作样本 —— 最少3份样本，或总字数不少于500字（优先级参考下方列表）
- 风格指南用途 —— AI训练？代笔者上岗培训？团队内容对齐？
- 权威范围 —— 有没有哪些话题他们希望显得更权威/更不权威？
- 已知禁用表述 —— 有没有他们已经明确想要避免的词汇或表达方式？
样本优先级（真实度从高到低）：
- 非正式的Slack消息或邮件（原始未编辑的表达）
- 播客或通话转录文本
- LinkedIn帖子或文章
- 网站文案（通常经过编辑，真实度较低）
最低样本门槛：如果样本总字数少于500字，请停止操作并提示：
"这些样本过短，无法提取可靠的特征。请补充2-3份其他样本——邮件、Slack消息或转录文本效果最佳。越随意、越未经打磨的样本越好。"
不要尝试从少于500字的样本中提取完整风格，可改为提供快速模式。
Phase 1: Sample Quality Assessment
第一阶段：样本质量评估
Before extracting, reason through:
- Sample authenticity: Are these samples from edited/polished contexts (website, press) or raw contexts (Slack, email)? More polish = less authentic voice.
- Sample variety: Do the samples cover different contexts (professional, casual, educational)? Single-context samples produce single-dimension voice guides.
- Exclusion check: Identify and flag patterns that are NOT the authentic voice:
- Platform formatting tics (LinkedIn line breaks, Twitter brevity forcing)
- Typos and autocorrect errors
- Phrases borrowed from others (quotes, retweets)
- Unusually formal writing (legal docs, press releases)
- Sample size adequacy: Is there enough material for full mode, or should I use quick mode?
Output a sample assessment:
"I have [X samples / Y words] to work with. Quality: [high/medium — why]. I'll use [full/quick] mode. Excluding: [any patterns and why]."
提取前请先判断以下几点：
- 样本真实度： 这些样本来自经过编辑打磨的场景（网站、新闻稿）还是原始场景（Slack、邮件）？打磨程度越高，风格真实度越低。
- 样本多样性： 样本是否覆盖了不同场景（专业、非正式、教育类）？单一场景的样本只会产出单维度的风格指南。
- 无效内容排查： 识别并标记不属于真实风格的特征：
- 平台格式特征（LinkedIn换行规则、Twitter字数限制导致的简短表达）
- 拼写错误和自动纠错导致的错误内容
- 借用他人的表述（引用、转发内容）
- 异常正式的写作内容（法律文档、新闻稿）
- 样本量充足度： 现有素材是否足够生成完整版本，还是应该使用快速模式？
输出样本评估结果：
"我获取了[X份样本 / Y字内容]。质量：[高/中等 —— 说明原因]。我将使用[完整/快速]模式。排除内容：[所有要排除的特征及原因]。"
Phase 2: Core Energy Extraction
第二阶段：核心表达特征提取
Identify the fundamental communication mode:
Role:
- Teacher (breaks things down systematically)
- Challenger (pushes back on assumptions)
- Cheerleader (builds confidence and momentum)
- Straight-shooter (cuts through BS efficiently)
Default energy:
- Calm authority ("Here's what works.")
- High enthusiasm ("This is exciting — let me show you.")
- Understated confidence ("I've seen this a hundred times.")
Recurring themes: What topics appear unprompted across samples? These are the things they actually care about.
识别基础的表达模式：
角色定位：
- 教育者（系统地拆解内容）
- 挑战者（质疑固有假设）
- 鼓励者（建立信心、推动行动）
- 直述者（高效跳过无效信息）
默认语气：
- 沉稳权威（"这就是有效的方案。"）
- 高度热情（"这太有意思了，我给你讲讲。"）
- 低调自信（"这种情况我见过几百次了。"）
高频主题： 样本中反复主动提及的话题有哪些？这些是用户真正关注的内容。
Phase 3: Phrase Extraction (Systematic)
第三阶段：表述提取（系统化）
Scan all samples and extract:
Transition phrases (how they shift topics):
- Quote exact examples from samples
- Pattern: "Here's the thing...", "What I've learned...", "Let me put it differently..."
Emphasis phrases (how they land a point):
- Quote exact examples
- Pattern: "The reality is...", "This is the part people miss...", "Here's the actual problem..."
Closers (how they wrap up):
- Quote exact examples
- Pattern: "That's the move.", "Start there.", "You've got this."
扫描所有样本，提取以下内容：
过渡短语（切换话题的方式）：
- 引用样本中的 exact 示例
- 示例模式："重点是...", "我学到的是...", "换个说法..."
强调短语（表达核心观点的方式）：
- 引用样本中的 exact 示例
- 示例模式："现实是...", "这是大家容易忽略的部分...", "真正的问题是..."
收尾短语（总结内容的方式）：
- 引用样本中的 exact 示例
- 示例模式："就这么做。", "从这里开始。", "你可以的。"
Phase 4: Confidence Zone Mapping
第四阶段：权威范围映射
| Zone | Description | Language Markers |
|---|---|---|
| Full authority | Topics they're an expert in | No hedging, definitive statements, "here's what works" |
| Earned perspective | Topics with experience but not mastery | "In my experience...", "What I've found..." |
| Active exploration | Topics they're learning now | "I'm testing this...", "What I'm seeing..." |
Map their stated expertise areas to each zone. This calibration is what makes the voice feel real vs. one-dimensional.
| 范围 | 描述 | 语言特征 |
|---|---|---|
| 完全权威 | 用户作为专家的话题 | 无模糊表述、明确的结论、"这就是有效的方案" |
| 经验视角 | 用户有经验但非精通的话题 | "根据我的经验...", "我发现..." |
| 探索阶段 | 用户正在学习的话题 | "我正在测试这个...", "我目前观察到..." |
将用户明确的专业领域对应到上述范围，这种校准可以让生成的风格更真实，而非单一刻板。
Phase 5: Anti-Pattern Documentation
第五阶段：反模式记录
Extract what they'd NEVER say:
- Words that would feel wrong in their voice
- Phrases that make them cringe
- Tones they naturally avoid
- Industry jargon they hate
Source these from sample evidence where possible: "You never used [word] across [X samples] — it doesn't fit your voice."
提取用户绝对不会使用的内容：
- 不符合其风格的词汇
- 让用户反感的短语
- 用户天然规避的语气
- 用户讨厌的行业黑话
尽可能用样本内容作为依据："你在[X份样本]中从未使用过[词汇]，它不符合你的表达风格。"
Phase 6: Validation Test (REQUIRED)
第六阶段：验证测试（必填）
After extracting the full profile, generate 2 test sentences on the same topic:
Version A (using the extracted voice profile):
"[Sample sentence in their voice]"
Version B (wrong voice — contrasting example):
"[Same content, different voice — shows what to avoid]"
Ask the user: "Does Version A actually sound like you when you're not overthinking it? What feels off?"
This validation catches extraction errors before the guide is put into production.
提取完完整的风格画像后，生成2条同主题的测试语句：
版本A（使用提取的风格画像）：
"[符合用户风格的示例语句]"
版本B（错误风格，作为对比）：
"[相同内容、不同风格的语句，展示需要规避的表达]"
询问用户："版本A听起来是不是你不加修饰时的真实表达？有没有哪里觉得不对？"
该验证可以在指南投入使用前排查提取错误。
Quick Mode (--quick
)
--quick快速模式（--quick
）
--quickWhen samples are thin (300–500 words) or time is short:
- Read 3 samples fast
- Pull 10 signature phrases
- Note 3 things they'd never say
- Write 1 sentence describing their energy
Output: Minimum viable voice guide.
Difference from full mode:
- Quick: ~10 phrases, 3 anti-patterns, 1-sentence energy descriptor
- Full: Complete profile with confidence calibration, validated test sentences, and source-cited examples
当样本量较少（300-500字）或者时间紧张时使用：
- 快速通读3份样本
- 提取10个标志性表述
- 记录3个用户不会使用的表述
- 用1句话描述用户的表达特征
输出： 最小可用风格指南
和完整模式的区别：
- 快速模式：约10个短语、3个反模式、1句话风格描述
- 完整模式：带权威校准、验证测试语句、来源标注示例的完整画像
Phase 7: Self-Critique Pass (REQUIRED)
第七阶段：自检环节（必填）
After generating the Voice Guide:
- Are the extracted phrases actually from the samples, or am I inferring them?
- Does the anti-pattern list include specific words/phrases, or just vague categories?
- Do the validation test sentences demonstrate a real difference between in-voice and out-of-voice?
- Is the confidence zone mapping specific to named topics, or just generic?
- Would a ghostwriter be able to use this guide without asking follow-up questions?
Flag any issues: "The anti-pattern section only has 2 entries — not enough for a usable guide. I need more samples or direct input from the user."
生成风格指南后，请检查：
- 提取的短语是否真的来自样本，还是自行推断的？
- 反模式列表是否包含具体的词汇/短语，还是只有模糊的分类？
- 验证测试语句是否真的能区分符合风格和不符合风格的表达？
- 权威范围映射是否对应具体的话题，还是通用模糊的描述？
- 代笔者是否可以直接使用这份指南，不需要额外询问？
标记所有问题："反模式部分只有2条内容，不足以作为可用指南，我需要更多样本或者用户的直接反馈。"
Output Structure
输出结构
markdown
undefinedmarkdown
undefinedVoice Guide: [Name] — [Date]
Voice Guide: [Name] — [Date]
Sample Assessment
Sample Assessment
- Samples: [count, types]
- Total words: [count]
- Quality: [high/medium — reason]
- Mode: [quick/full]
- Excluded: [patterns excluded + why]
- Samples: [count, types]
- Total words: [count]
- Quality: [high/medium — reason]
- Mode: [quick/full]
- Excluded: [patterns excluded + why]
Core Energy
Core Energy
- Role: [teacher/challenger/cheerleader/straight-shooter]
- Default energy: [description]
- Recurring themes: [list]
- Role: [teacher/challenger/cheerleader/straight-shooter]
- Default energy: [description]
- Recurring themes: [list]
Signature Phrases
Signature Phrases
Transitions:
- "[Phrase]" (source: [email/post])
- "[Phrase]"
Emphasis:
- "[Phrase]" (source: [email/post])
Closers:
- "[Phrase]"
Transitions:
- "[Phrase]" (source: [email/post])
- "[Phrase]"
Emphasis:
- "[Phrase]" (source: [email/post])
Closers:
- "[Phrase]"
Confidence Calibration
Confidence Calibration
Full authority (no hedging):
Topics: [list]
Sounds like: "[example sentence]"
Earned perspective:
Topics: [list]
Sounds like: "[example sentence]"
Active exploration:
Topics: [list]
Sounds like: "[example sentence]"
Full authority (no hedging):
Topics: [list]
Sounds like: "[example sentence]"
Earned perspective:
Topics: [list]
Sounds like: "[example sentence]"
Active exploration:
Topics: [list]
Sounds like: "[example sentence]"
Anti-Patterns (Never Use)
Anti-Patterns (Never Use)
- [Word/phrase] — why: [evidence from samples]
- [Word/phrase] — why: [evidence]
- [Word/phrase] — why: [evidence from samples]
- [Word/phrase] — why: [evidence]
Validation Test
Validation Test
This sounds like you:
"[Version A]"
This doesn't:
"[Version B — contrast]"
This sounds like you:
"[Version A]"
This doesn't:
"[Version B — contrast]"
Self-Critique Notes
Self-Critique Notes
[Any gaps, things to validate with user]
[Any gaps, things to validate with user]
Usage Instructions
Usage Instructions
- For AI: Paste this guide into your system prompt
- For ghostwriter: Share on day 1 — cuts revision cycles in half
- For team: This is the benchmark for "on brand"
---
*Skill by Brian Wagner | AI Marketing Architect | brianrwagner.com*- For AI: Paste this guide into your system prompt
- For ghostwriter: Share on day 1 — cuts revision cycles in half
- For team: This is the benchmark for "on brand"
---
*Skill by Brian Wagner | AI Marketing Architect | brianrwagner.com*