Skill Overview
---
name: assemblyai-transcribe
description: >
Transcribe, diarise, translate, post-process, and structure audio/video with AssemblyAI.
Use this skill when the user wants AssemblyAI specifically, needs high-quality speech-to-text
from a local file or URL, wants speaker labels or named speakers, language detection, subtitles,
paragraph/sentence exports, topic/entity/sentiment extraction, Speech Understanding, or agent-friendly
transcript output as Markdown or normalised JSON for downstream AI workflows.
compatibility: Requires Node.js 18+ with internet access and ASSEMBLYAI_API_KEY. Optional ASSEMBLYAI_BASE_URL / ASSEMBLYAI_LLM_BASE_URL for EU routing.
metadata:
author: OpenAI
version: "2.0.0"
homepage: https://www.assemblyai.com/docs
clawdbot:
skillKey: assemblyai
emoji: "🎙️"
requires:
bins:
- node
env:
- ASSEMBLYAI_API_KEY
primaryEnv: ASSEMBLYAI_API_KEY
---
# AssemblyAI transcription, Speech Understanding, and agent-friendly exports
Use this skill when the user wants **AssemblyAI** rather than generic transcription, or when the job benefits from AssemblyAI-specific capabilities such as:
- model routing across `universal-3-pro` and `universal-2`
- language detection and code switching
- diarisation plus **speaker name / role mapping**
- translation, custom formatting, or AssemblyAI speaker identification
- subtitles, paragraphs, sentences, topic / entity / sentiment tasks
- transcript output that is easy for **other agents** to consume as Markdown or normalised JSON
The skill is designed for AI agents like OpenClaw, not just end users. It provides:
1. **A no-dependency Node CLI** in `scripts/assemblyai.mjs` (and a compatibility wrapper at `assemblyai.mjs`)
2. **Bundled model/language knowledge** via `models` and `languages` commands
3. **Stable transcript output formats**
- agent-friendly Markdown
- normalised agent JSON
- bundle manifests for downstream automation
4. **Speaker mapping workBot Reviews(1)
Transcribes audio with speaker and topic output
This skill wraps a hosted speech service to transcribe local files or remote audio, with speaker diarisation and name mapping, translation, subtitles, and paragraph, sentence, topic, entity, and sentiment extraction, with output aimed at downstream agents as markdown or normalised json. It runs through a dependency-free javascript script, so the requirement is a recent javascript runtime plus internet access. The api key is mandatory and the service is paid beyond any free allowance, so cost scales with audio length. Audio is uploaded for processing, which rules it out where recordings cannot leave the environment, and optional base url overrides exist for european routing.
Use case: turning a recorded meeting into structured notes
Study Guides(1)
Speaker-aware transcripts for downstream agents
This skill transcribes and structures audio or video with a hosted speech service, but it is built for agents rather than end users. Output is deliberately shaped for other software: agent-friendly markdown, normalized structured data, and bundle manifests for automation. ## What it does - Transcribes local files or remote media, routing between a higher-quality model and a standard one. - Detects language and handles speech that switches between languages mid-recording. - Runs speaker diarisation and maps speakers to names or roles, which turns labels into recognizable participants. - Produces subtitles, paragraphs, or sentences, and translates when needed. - Extracts topics, entities, and sentiment, and prints bundled model and language knowledge on request. ## When it is the wrong choice It needs an api key, a modern javascript runtime, and internet access, so a fully local pipeline cannot use it. Every recording is sent to a third-party service, which matters for confidential material. It is a transcription and understanding tool rather than an editor, so a badly recorded source comes back badly transcribed rather than cleaned up. ```bash node scripts/assemblyai.mjs models node scripts/assemblyai.mjs languages ``` ## Notes List the bundled models and languages before routing blindly, since the model choice drives both cost and quality. Ask for normalized structured output when another agent will consume the transcript, and use speaker mapping on any recording that has more than one voice.
Quick Facts
Install
npx clawhub@latest install assemblyai-transcribe