菜单
腾讯云语音识别 ASR skill，用于语音转文字、音频转写、字幕生成、会议转录、语音消息识别、本地文件或 URL 音频识别。覆盖短音频、中长音频快速识别和长音频异步识别，支持普通话、英语、粤语、日语、韩语、德语、中英粤混说和多种中文方言。
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
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。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/whyzsm/tiny-agents --skill tencentcloud-asr命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Single-Agent T-MAX Java backend architecture and engineering rule package. Use when Codex must design, implement, review, or diagnose T-MAX backend services involving Java 8, Spring Boot, Kylin, MyBatis XML, TiDB/MySQL, Apollo/Nacos, Feign, AnePageHelper, ResponseResult/DataGridResult, UEP export, Redis, RabbitMQ, or Maven multi-module repositories. Routes one Agent through repository intake, architecture decisions, API/data contracts, implementation diagnosis, quality/security review, and bounded evidence loops while keeping builds, tests, live calls, commits, and pushes as separate authority boundaries.
Create polished consulting-style single-file HTML research reports from messy business notes, screenshots, metrics, and iterative user feedback. Use this skill whenever the user asks for a 咨询风/艾瑞咨询风/研究报告/汇报报告/HTML报告, wants business or AI landing-practice material turned into a management-facing report, asks to add charts/diagrams, or asks to update report metrics and keep the whole HTML report consistent.
Audit HarmonyOS applications for AppGallery/AGC release readiness, including Huawei UX/HMI controls, system-feature integrations, and performance gates for multi-input interaction, latency, frame rate, content display, memory, and CPU. Use when users ask for 鸿蒙上架自检、应用市场审核、AGC发布前检查、控件/系统特性/人机交互/签名/隐私/权限/截图/性能核对或 release 包门禁, especially for Stage-model projects with module.json5, build-profile.json5, hvigorw, and local-first data constraints.
基于 SOC 职业分类
正在显示 SKILL.md
| name | tencentcloud-asr |
| description | 腾讯云语音识别 ASR skill，用于语音转文字、音频转写、字幕生成、会议转录、语音消息识别、本地文件或 URL 音频识别。覆盖短音频、中长音频快速识别和长音频异步识别，支持普通话、英语、粤语、日语、韩语、德语、中英粤混说和多种中文方言。 |
腾讯云语音识别（ASR），微信同款ASR引擎，历经亿级用户场景验证，稳定可靠。在中英混说场景下识别效果行业领先，精准流畅。支持普通话、方言及多语种识别，提供一句话识别、录音识别等全场景能力，是高性价比语音转文字首选。
inspect_audio.pysentence_recognize.py、flash_recognize.py 或 file_recognize.pyself_check.pypython3 <SKILL_DIR>/scripts/inspect_audio.py "<AUDIO_INPUT>"。ffmpeg / ffprobe 先自治安装：先执行 python3 <SKILL_DIR>/scripts/ensure_ffmpeg.py --execute，只有失败后才向用户求助。python3 <SKILL_DIR>/scripts/self_check.py，不要先让用户手工试脚本。SecretId、SecretKey、AppId~/.bashrc、~/.zshrc对话语言只能当作先验，不等于音频语种本身。若用户音频语种明显不同，按音频语种改。
| 场景 | 一句话识别 | 极速版 | 录音文件识别 | 备注 |
|---|---|---|---|---|
| 普通话 | 16k_zh | 16k_zh / 16k_zh_large | 16k_zh / 16k_zh_large | 默认首选 |
| 中英夹杂 | 16k_zh-PY | 16k_zh_en | 16k_zh_en | 混说优先 |
| 粤语 | 16k_yue | 16k_yue | 16k_yue | |
| 英语 | 16k_en | 16k_en | 16k_en / 16k_en_large | |
| 日语 | 16k_ja | 16k_ja | 16k_ja | |
| 韩语 | 16k_ko | 16k_ko | 16k_ko | |
| 多语种 / 语言不确定 | 指定具体语种 | 16k_multi_lang | 16k_multi_lang | 一句话识别没有多语自动识别引擎 |
如果有多个明显可选项：
16kHz、单声道、pcm_s16le、.wav<=60s 且 <=3MB：sentence_recognize.py<=2h 且 <=100MB：优先 flash_recognize.py<=5h，可走 file_recognize.py recfile_recognize.py recfile_recognize.py rec 真实失败时，再按错误决定是否进入本地下载 / 规范化 / 切片链命中 URL、大文件、切片、body vs URL 取舍时，再读 routing_strategy.md。
# 预检
python3 <SKILL_DIR>/scripts/inspect_audio.py "<AUDIO_INPUT>"
# 凭证自检
python3 <SKILL_DIR>/scripts/self_check.py
# 一句话识别
python3 <SKILL_DIR>/scripts/sentence_recognize.py "<AUDIO_INPUT>" --engine 16k_zh
# 极速版
python3 <SKILL_DIR>/scripts/flash_recognize.py "<AUDIO_INPUT>" --engine 16k_zh
# 录音文件识别
python3 <SKILL_DIR>/scripts/file_recognize.py rec "<AUDIO_INPUT_OR_URL>" --engine 16k_zh
# CLI transcription backend
python3 <SKILL_DIR>/scripts/cli_transcribe.py "<MEDIA_PATH_OR_URL>"
scripts/inspect_audio.py：音频探测scripts/ensure_ffmpeg.py：自治安装 ffmpeg / ffprobescripts/self_check.py：凭证与三种模式自检scripts/sentence_recognize.py：一句话识别scripts/flash_recognize.py：录音文件识别极速版scripts/file_recognize.py：录音文件识别异步任务scripts/cli_transcribe.py：CLI backend wrapper