一键在 Manus 中运行任何 Skill

mimo-tts-wav

星标1

分支0

更新时间2026年7月2日 10:53

Generate a spoken or sung WAV audio file using MiMo TTS. USE THIS SKILL whenever the user asks the model to speak, say something aloud, read text out, sing a song, or produce any audio/voice output — even if no explicit destination is mentioned. Save the result to the current working directory as a .wav file (default name: output.wav) unless the user specifies a different path. Supports any expressive style and custom voice cloning.

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

Jinhong2706

Jinhong2706/AI-Agent

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

相关职业SOC

基于 SOC 职业分类

软件开发工程师计算机与数学类职业·SOC 15-1252

文件资源管理器

3 个文件

SKILL.md

readonly

同仓库更多 Skills

同仓库

setup-writing-style

Jinhong2706/AI-Agent

Learns how the user writes from their own sent messages and docs, and builds a voice profile so future drafts sound like them instead of generic AI. The profile is saved as the my-writing-style skill. Use when the user asks to set up, learn, or capture their writing voice, or complains that drafts sound generic or unlike them and no my-writing-style profile exists. Only for drafting text the user will send as themselves, not for Claude's own replies.

2026-07-111

docx

Jinhong2706/AI-Agent

Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files) or Word templates (.dotx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', '.dotx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx or .dotx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.

2026-07-111

pptx

Jinhong2706/AI-Agent

Use this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates (.potx), layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx or .potx filename, regardless of what they plan to do with the content afterward. If a .pptx or .potx file needs to be opened, created, or touched, use this skill.

2026-07-111

xlsx

Jinhong2706/AI-Agent

Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.

2026-07-111

image-gen-edit

Jinhong2706/AI-Agent

Use this skill when replying with a generated or edited image — when a post asks to create, generate, edit, modify, restyle, combine, or stitch images, or whenever attaching a generated or edited image would strengthen the reply.

2026-07-101

doubao-app-builder

Jinhong2706/AI-Agent

统一处理网页应用的生成、编辑，以及围绕已生成产物的问答。既负责把自然语言需求端到端转成可运行、可预览、可交付的网页应用产物，也负责在用户追问产物时基于真实产物作答。当用户要生成网站、H5、网页应用、管理后台、数据看板时使用。当用户要编辑已有网页应用、做功能新增、页面调整或 Bug 修复时使用。当用户提供 PRD、文档、截图或素材包并要求产出可预览网页应用时使用。当用户针对已生成的网页应用，要求总结或解读网页内容、查看或分析源码、解释或排查运行报错、查询访问量/用户量/数据报表/发布状态等运营信息时同样使用。

2026-07-021

name	mimo-tts-wav
description	Generate a spoken or sung WAV audio file using MiMo TTS. USE THIS SKILL whenever the user asks the model to speak, say something aloud, read text out, sing a song, or produce any audio/voice output — even if no explicit destination is mentioned. Save the result to the current working directory as a .wav file (default name: output.wav) unless the user specifies a different path. Supports any expressive style and custom voice cloning.
applyTo	**

MiMo TTS → WAV File

Convert text to speech using the MiMo TTS API and save the result as a standard WAV file (24 kHz, 16-bit PCM, mono) to any local path you choose.

When to Use

Situation	Use this skill?
User says "说..." / "读出来" / "大声念" / "speak" / "say"	✅ Yes
User says "唱..." / "唱个歌" / "唱歌" / "sing"	✅ Yes
User asks for any voice/audio output	✅ Yes
No output path given — save to `$PWD/output.wav`	✅ Yes
Need to send voice directly to Feishu	❌ Use `mimo-tts-voice` instead
Need to post-process audio (ffmpeg, SoX …)	✅ Yes — pipe the output

Prerequisites

Requirement	Details
`jq`	JSON builder. Install: `brew install jq`
`python3`	Standard library only (json, base64, struct). Pre-installed on macOS.
`curl`	Pre-installed on macOS.

Usage

bash scripts/tts_to_wav.sh "<text>" "<output.wav>" [style] [voice_sample.wav]

Arguments

#	Name	Required	Description
1	`text`	✅	Text to synthesize. Apply TTS normalization rules before passing. Singing mode: pass a song name (looked up in `sing0301_dict.json`) or `LYRICS:<your lyrics>` to supply lyrics directly.
2	`output_path`	✅	Destination path for the WAV file. Parent directory must exist.
3	`style`	optional	Any natural-language style phrase (see Style Guide below). Pass `唱歌` to enable singing mode.
4	`voice_sample`	optional	Path to a WAV reference clip (24 kHz, 16-bit, mono, 5–15 s) for voice cloning. Omit to use the preset voice `mimo_default`.

Environment Variables

Variable	Default	Description
`MIMO_VOICE_SAMPLE`	—	Fallback voice sample path (used when `voice_sample` arg is omitted)
`MIMO_API_BASE_URL`	`https://api.xiaomimimo.com/v1`	Override API base URL (`/chat/completions` is appended automatically)
`MIMO_TTS_MODEL`	`mimo-v2.5-tts`	Override model name
`MIMO_SING_DICT`	`../../sing0301_dict.json` (relative to script)	Path to `sing0301_dict.json` song lyrics dictionary

Step-by-Step

Normalize the text — Apply TTS normalization (see section below) so numbers, symbols and formatting render as natural speech.
Choose a voice
- Preset: omit voice_sample → uses mimo_default
- Custom clone: pass a 5–15 s WAV clip as the 4th argument, or set MIMO_VOICE_SAMPLE
Choose a style (optional) — pass any descriptive phrase as the 3rd argument. Omit for neutral speech.

Run the script:

bash scripts/tts_to_wav.sh "要说的内容" "/path/to/out.wav"

Check the output — the script exits 0 and prints OK: <N> bytes written to <path> on success. Any error is printed to stderr with a non-zero exit code.

Examples

# Minimal — preset voice, no style
bash scripts/tts_to_wav.sh "今天天气不错" /tmp/hello.wav

# With a mood style
bash scripts/tts_to_wav.sh "恭喜你获得了一等奖！" /tmp/congrats.wav "开心激动"

# Slow delivery for an accessibility recording
bash scripts/tts_to_wav.sh "请系好安全带。" /tmp/safety.wav "语速慢 清晰"

# Custom voice clone + style
bash scripts/tts_to_wav.sh "大家好，我是小明。" /tmp/xiaoming.wav "热情" ~/voices/xiaoming.wav

# Singing — look up lyrics by song name (requires sing0301_dict.json)
bash scripts/tts_to_wav.sh "两只老虎" /tmp/song.wav "唱歌"

# Singing — supply lyrics directly with LYRICS: prefix
bash scripts/tts_to_wav.sh "LYRICS:两只老虎，两只老虎，跑得快，跑得快……" /tmp/song.wav "唱歌"

# Pipe straight into ffmpeg to get MP3
bash scripts/tts_to_wav.sh "你好" /tmp/tmp.wav && \
  ffmpeg -i /tmp/tmp.wav /tmp/out.mp3

Style Guide

style accepts any natural-language phrase — the model interprets it freely. There are no fixed values.

Category	Examples
Emotion	`开心` / `悲伤` / `生气` / `平静`
Delivery	`语速慢` / `语速快` / `悄悄话` / `清晰有力`
Character	`像个大将军` / `像个小孩` / `孙悟空` / `林黛玉`
Dialect	`东北话` / `四川话` / `台湾腔` / `粤语`
Combinations	`慵懒刚睡醒` / `撒娇夹子音` / `深情款款语速慢`
Singing	`唱歌` — text is a song name (looked up in `sing0301_dict.json`) or `LYRICS:<lyrics>`

Omit the style argument entirely for neutral, natural-sounding speech.

TTS Text Normalization

Preprocess the input text before passing it to the script. Replace unspoken symbols with natural spoken Chinese (or English where appropriate).

Numbers

Input	Spoken
`3`	三
`3.14`	三点一四
`1/3`	三分之一
`14:30`	下午两点半
`95%`	百分之九十五
`2024` (year)	二零二四年

Common Symbols

Symbol	Spoken
`+`	加
`-`	减
`×` / `*`	乘以
`÷` / `/`	除以
`=`	等于
`>` / `<`	大于 / 小于
`%`	百分之…
`~`	大约
`...` / `…`	等等
`#`	井号 (or 第, by context)

Formatting Artifacts

Remove markdown (**bold**, # heading, `code`, - bullet)
Convert numbered lists to prose: "有三点，第一……第二……第三……"
Convert tables to sentences describing key values

Output Format

The script always writes a standard RIFF WAV file:

Property	Value
Format	WAV (RIFF)
Sample rate	24 000 Hz
Bit depth	16-bit PCM
Channels	Mono

If the API returns raw PCM instead of WAV, the script automatically adds the correct RIFF header before writing to disk.

Error Reference

Exit code	Meaning	Fix
`1`	Missing required argument	Provide `text` and `output_path`
`1`	Voice sample file not found	Check the path passed as arg 4
`2`	API HTTP error	Check network access; API key is resolved automatically
`2`	Unexpected response shape	Inspect stderr for raw API response
`3`	Output file empty / not written	Check that the parent directory exists and is writable