Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add bleakbelladonnals/asr-transcription-skills --skill asr-benchmarkinggit clone --depth 1 https://github.com/bleakbelladonnals/asr-transcription-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/bleakbelladonnals/asr-transcription-skills/asr-benchmarking)<a href="https://agentmods.dev/skills/bleakbelladonnals/asr-transcription-skills/asr-benchmarking"><img src="https://agentmods.dev/badge/skills/bleakbelladonnals/asr-transcription-skills/asr-benchmarking/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/bleakbelladonnals/asr-transcription-skills/asr-benchmarking"><img src="https://agentmods.dev/badge/skills/bleakbelladonnals/asr-transcription-skills/asr-benchmarking.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00120 | $0.01984 |
| Opus 5 | $0.00060 | $0.00992 |
| Sonnet 5 | $0.00024 | $0.00397 |
| Haiku 4.5 | $0.00012 | $0.00198 |
Grade A, and why
asr-benchmarking scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ASR 转写评测与多平台 API 集成
触发场景
- 用户问"录音转文字做的最好的模型/产品是什么"
- 对比录音硬件/云转写产品的音频质量、转写质量、纪要质量、成本、生态
- 把同一段音频批量交给多个转写引擎（Gemini、百炼、火山、OpenAI、剪映）做横向对比
- 为评测文章/课程作业准备可溯源数据
核心方法论
1. 双变量拆分（评测设计灵魂）
"端到端质量"= 采集端 × 转写端，必须拆开测：
- 采集端（硬件麦克风/DSP/AGC）：同期多设备同位置录制同一内容，比音频文件本身
- 转写端（模型/引擎）：同一音频文件喂不同引擎，比转写结果 混着测只能得出"谁家成品好"，说不清"是硬件好还是模型好"。
2. 音频质量客观指标（ffmpeg，macOS 自带 ffprobe）
# 基本参数：编码/码率/采样率/声道/时长
ffprobe -v quiet -show_format -show_streams f.mp3 | grep -E "duration|bit_rate|sample_rate|channels|codec_name"
# 响度与动态范围（前5分钟即可）
ffmpeg -v info -i f.mp3 -t 300 -af ebur128 -f null - 2>&1 | grep -A8 Summary # I=LUFS 响度, LRA=动态范围
# 电平/削波：Peak >0dB = 削波风险
ffmpeg -v info -i f.mp3 -af astats -f null - 2>&1 | grep -E "RMS level|Peak level|DC offset"
# 静音段检测：-38dB 阈值找不到静音 = AGC 抬底噪
ffmpeg -v info -i f.mp3 -af silencedetect=noise=-38dB:d=2.5 -f null - 2>&1 | grep silence_
# 多声道是否同源：两声道 RMS 几乎相等 = 非真立体声，转写无额外信息
ffmpeg -v info -ss 2160 -t 30 -i f.ogg -af astats -f null - 2>&1 | grep -E "Channel|RMS level"
解读要点：码率差（128k MP3 vs 65k Opus）≠ 音质差一半（Opus 效率高 2 倍）；LRA 小 = 压缩狠；Peak 超 0dB = 削波；"找不到静音"= 底噪被抬升。实测例：钉钉录音卡动态 13.9 LU/有静音段 vs 飞书录音豆 8.2 LU/无静音段/Peak 1.38dB 削波风险。
3. 转写质量对比（无金标准也能比）
- 结构指标：块数、总字数、片段均长/中位、<1s 占比（碎片化）、每分钟字数、说话人标记
- 术语锚点对齐（最强武器）：选 3-5 句含专有名词的话做"固定测试集"，每份转写查同一句。实例：同一句"…只能用 Claude Code"，钉钉听记 46:14 全对，飞书妙记 46:25 错成 "Clash code"——铁证级对比
- 幻觉检查：凭空英文/无意义字母（AN、DK）/误听脏话（"我他妈"）
- AI 纪要核验：逐字稿对照纪要每条事实，标出"合理推测被写成事实"（如"回滚功能""定时触发"逐字稿根本没有）
- 对齐方式：按内容锚点对齐，不能按时间戳对齐——不同设备录音时间轴会错位（智能暂停/启动时间差，实测前段错位 4-17 分钟、后段对齐 2-3 秒）
4. 多平台 API 转写（统一策略）
- 所有平台统一喂 16kHz 单声道 mp3 分段（ffmpeg 转）：一次解决 OGG 不兼容（讯飞踩坑）、25MB 限制、token 限额
- 默认 15 分钟/段（≈2.4 万 token），72 分钟课 = 5 段
- prompt 强制逐字：保留口语/重复/语气词，英文术语原样，temperature=0
- 模型 ID 时效性铁律：API 模型不是永久可用（gemini-2.5-flash 对 2026 年新用户返回 404 "no longer available to new users"）。确定模型 ID 的顺序：模型列表 API/平台控制台 > 官方最新文档 > 记忆。调不通先列模型列表（Gemini: GET /v1beta/models?key=）
- 先读官方最新文档再定模型（用户明确要求的工作流）
5. 交付形态（用户偏好）
- 用户不写代码：脚本必须零第三方依赖（标准库 + ffmpeg），key 用同目录 key 文件（gemini_key.txt / bailian_key.txt / volc_key.txt，粘贴即用），不要以环境变量/装包教程为主
- 报告：中文、结论先行、表格化、数据可溯源（每行标注来源）
- 成本敏感：优先免费额度与按量 API，给出每小时的账
- 评测结论要能直接改写成公众号/人人都是产品经理文章（用户常需素材包+大纲）
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 71 lines · 120 tokens per session scan A 2314d61fe6f1
asr-benchmarking is a skill published in the GitHub repository bleakbelladonnals/asr-transcription-skills (9 stars, last pushed 26d ago), licensed MIT. It adds 120 tokens to every session and 1,984 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…