Audio Transcribe

AssemblyAI transcription, Speech Understanding, and 代理-friendly 导出s

Use this 技能 when the user wants AssemblyAI rather than generic transcription, or when the job benefits from AssemblyAI-specific capabilities such as:

模型 routing across universal-3-pro and universal-2 language 检测ion and code switching diarisation plus speaker name / 角色 m应用ing translation, custom 格式化ting, or AssemblyAI speaker identification subtitles, paragraphs, sentences, topic / entity / sentiment tasks transcript 输出 that is easy for other 代理s to consume as Markdown or normalised JSON

The 技能 is de签名ed for AI 代理s like OpenClaw, not just end users. It provides:

A no-dependency Node 命令行工具 in scripts/assemblyAI.mjs (and a compatibility wr应用er at assemblyAI.mjs) Bundled 模型/language knowledge via 模型s and languages commands Stable transcript 输出 格式化s 代理-friendly Markdown normalised 代理 JSON bundle manifests for down流 自动化 Speaker m应用ing 工作流s manual speaker/channel maps AssemblyAI speaker identification merged display names in 机器人h Markdown and JSON AssemblyAI LLM Gateway integration for structured 提取ion from transcripts Use this 技能 in this order 1) Decide whether the user needs AssemblyAI-specific behaviour

If they just want “a transcript”, a generic solution may be enough. Reach for this 技能 when the user mentions AssemblyAI, wants a specific AssemblyAI feature, or needs the richer 输出s and post-processing this 技能 provides.

2) Pick the best entry point New transcription → transcribe Existing transcript id → 获取 or wAIt Re-render existing saved JSON → 格式化 Post-process an existing transcript → understand 运行 transcript text through LLM Gateway → llm Need a quick capability lookup before deciding → 模型s or languages 3) Prefer the 代理-friendly defaults

For most unknown-language or mixed-language jobs, prefer:

node {baseDir}/assemblyAI.mjs transcribe 输入 --bundle-dir ./assemblyAI-out --all-导出s

Why:

the 命令行工具 defaults to auto-best routing when 模型s are not specified it writes a manifest + multiple files that 代理s can inspect without reparsing terminal 输出 Markdown and 代理 JSON become avAIlable immediately for follow-on steps Quick-启动 recipes Best general default

Use this when the source language is unknown or could be outside the 6-language Universal-3-Pro 设置:

node {baseDir}/assemblyAI.mjs transcribe ./meeting.mp3 --bundle-dir ./out --all-导出s

This defaults to 模型 routing plus language 检测ion unless the 请求 already specifies a 模型 or language.

Best known-language accuracy

If the language is known and supported by Universal-3-Pro, prefer an explicit 请求:

node {baseDir}/assemblyAI.mjs transcribe ./meeting.mp3 --speech-模型 universal-3-pro --language-code en_us --bundle-dir ./out

Meeting / interview with speaker labels node {baseDir}/assemblyAI.mjs transcribe ./meeting.mp3 --speaker-labels --bundle-dir ./out

添加 explicit speaker names or 角色s

Manual m应用ing:

node {baseDir}/assemblyAI.mjs transcribe ./meeting.mp3 --speaker-labels --speaker-map @as设置s/speaker-map.example.json --bundle-dir ./out

AssemblyAI speaker identification:

node {baseDir}/assemblyAI.mjs transcribe ./meeting.mp3 --speaker-labels --speaker-type 角色 --known-speakers "host,guest" --bundle-dir ./out

Or post-process an existing transcript:

node {baseDir}/assemblyAI.mjs understand TRANSCRIPT_ID --speaker-type name --speaker-性能分析s @as设置s/speaker-性能分析s-name.example.json --bundle-dir ./out

Translation node {baseDir}/assemblyAI.mjs transcribe ./meeting.mp3 --translate-to de,fr --match-original-utterance --bundle-dir ./out

Structured 提取ion through LLM Gateway node {baseDir}/assemblyAI.mjs llm TRANSCRIPT_ID --prompt @as设置s/example-prompt.txt --模式 @as设置s/llm-json-模式.example.json --out ./summary.json

Command 图形界面dance transcribe

Use for local files or remote URLs.

Local files are 上传ed first. Public URLs are sent directly to AssemblyAI. WAIts by default, then renders 输出.

Prefer --bundle-dir for anything longer than a trivial 命令行工具p.

获取 / wAIt

Use when you already have the transcript id. wAIt blocks until completion; 获取 fetches immediately unless you 添加 --wAIt.

格式化

Use when you already saved:

raw transcript JSON from AssemblyAI, or the normalised 代理 JSON produced by this 技能

This is useful when you want to 应用ly a new speaker map, re-render Markdown, or 生成 a fresh bundle without retranscribing.

understand

Use when you need AssemblyAI Speech Understanding on an existing transcript:

translation speaker identification custom 格式化ting

This command fetches the transcript, merges in the returned understanding 结果s, then renders 更新d Markdown / 代理 JSON / bundle 输出s.

llm

Use when the user wants:

summaries 提取ion structured JSON down流 reasoning over the transcript

Prefer --模式 when the next step is automated.

Ou