MAI Transcribe for Openclaw

High-accuracy audio transcription via Microsoft's MAI-Transcribe-1 model integrated through Azure AI Speech.

robotsbuildrobots
v0.1.1
Apr 7, 2026
0
668
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install mai-transcribe

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install mai-transcribe using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is MAI Transcribe?

MAI Transcribe is a specialized utility designed for developers who need reliable speech-to-text capabilities within their Openclaw Skills workflows. By leveraging Microsoft's advanced MAI-Transcribe-1 model, this skill provides high-fidelity transcriptions for various audio formats like M4A, OGG, and WAV.

Unlike generic tools, it focuses on direct integration with Azure AI Speech resources, ensuring enterprise-grade performance and security. This skill acts as a lightweight wrapper that bridges local audio processing with cloud-based intelligence, making it an essential addition for anyone building media-rich Openclaw Skills applications that require precise transcription.

MAI Transcribe Use Cases

  • Converting meeting recordings or voice notes into searchable text documents.
  • Automating transcription workflows for podcasts or video voiceovers.
  • Extracting raw JSON metadata for advanced NLP analysis and sentiment tracking.
  • Building localized transcription services for international projects using specific language flags.

How MAI Transcribe Works

  1. The user initiates the process by providing an audio file path to the transcription script via the CLI.
  2. The skill validates the required Azure environment variables, specifically the speech endpoint and API key.
  3. The audio file is packaged into a multipart form request along with a specific model definition payload.
  4. The request is transmitted to the configured Azure AI Speech endpoint using the 2025-10-15 API version.
  5. Azure's MAI-Transcribe-1 model processes the audio stream to generate text or structured JSON.
  6. The resulting transcription is saved locally to a file or output to the console for further integration within Openclaw Skills pipelines.

MAI Transcribe Setup

To use this skill, ensure you have Node.js installed and set up your Azure Speech credentials:

# Export your Azure Speech credentials
export AZURE_SPEECH_ENDPOINT="https://YOUR-RESOURCE.cognitiveservices.azure.com"
export AZURE_SPEECH_KEY="YOUR_SPEECH_RESOURCE_KEY"

# Run the transcription script
node scripts/transcribe.js /path/to/audio.m4a

MAI Transcribe Data Schema & Taxonomy

The skill handles data through CLI arguments and environment variables, producing clean text or structured JSON output.

Parameter Description Default
--out Custom path for the transcription output <input_file>.txt
--language Specify the BCP-47 language code (e.g., en-GB) Auto-detected
--json Returns the raw Azure API response for debugging False
--model Selection of the transcription model mai-transcribe-1

MAI Transcribe Advanced Features

  • Support for multiple audio containers including WAV, M4A, and OGG.
  • Enhanced Mode integration for higher accuracy using the MAI-Transcribe-1 engine.
  • Raw JSON export mode for developers needing access to word-level timestamps or confidence scores.
  • Flexible API versioning support via the AZURE_SPEECH_API_VERSION environment variable.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*