← Back to Skills Marketplace
182
Downloads
0
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install burmese-audio-understanding
Description
High-accuracy Burmese audio transcription using Gemini 3.1 Flash Preview.
README (SKILL.md)
Burmese Audio Understanding Skill
This skill allows you to transcribe Burmese audio (voice notes, speech) directly into Burmese text using your own Google Gemini API key. It uses the official Google GenAI SDK for secure and reliable file handling.
Required Environment Variables
GEMINI_API_KEY: Required. Set your Google Gemini API key to allow the skill to access transcription services.
Usage
Ensure GEMINI_API_KEY is set in your environment, then run:
node scripts/transcribe-direct.js /path/to/my-audio.ogg
Features
- Official SDK: Uses the official
@google/genaiSDK. - Improved Security: No shell commands (ffmpeg/child_process) used; file processing is handled via SDK file upload directly to Gemini.
- Model: Uses
gemini-3.1-flash-previewfor high-quality audio transcription.
Security Notes
- This skill sends audio data to Google Gemini API for transcription.
- No data is stored locally after processing.
- Requires a valid GEMINI_API_KEY with minimal permissions.
Prerequisites
- Dependencies must be installed:
npm install @google/genai.
Usage Guidance
This skill appears to do exactly what it says: upload a local audio file to Google Gemini for transcription. Before installing or running it: ensure you trust the @google/genai package version you'll install, protect the GEMINI_API_KEY (store it securely and grant only necessary permissions), be aware that audio data is transmitted to Google's servers (including any personally identifying content), and revoke the key if it is ever exposed. Also consider running npm installs in an isolated environment (container or VM) and confirm the package.json version matches the published registry artifact if version provenance matters to you.
Capability Analysis
Type: OpenClaw Skill
Name: burmese-audio-understanding
Version: 1.2.2
The skill is a legitimate implementation for Burmese audio transcription using the official @google/genai SDK. The script 'scripts/transcribe-direct.js' follows documented procedures for uploading audio files to the Gemini API, generating content, and performing cleanup by deleting the uploaded file. There are no signs of data exfiltration, shell execution, or malicious prompt injection.
Capability Assessment
Purpose & Capability
Name/description, SKILL.md, package.json, and the provided script all focus on sending a local audio file to the Google Gemini API for transcription. The declared requirement (GEMINI_API_KEY) matches the stated integration with Google Gemini. Minor: package.json version (1.2.1) differs from registry metadata (1.2.2), but this is a packaging/version inconsistency rather than a functional mismatch.
Instruction Scope
Runtime instructions are limited and specific: set GEMINI_API_KEY, run the Node script on a local audio file. The script uploads the file via the SDK, requests a transcription, prints it, and deletes the uploaded file. It does not reference unrelated files, extra environment variables, or external endpoints beyond the Google Gemini API.
Install Mechanism
There is no install spec; dependencies are standard npm (the SKILL.md requests running `npm install @google/genai`). Installing a single official SDK dependency is proportional to the task. No downloads from untrusted URLs or archive extraction are present.
Credentials
Only GEMINI_API_KEY is required and is appropriate for calling the Google Gemini API. No unrelated secrets or system credential paths are requested.
Persistence & Privilege
The skill is not always-enabled, does not request persistent elevated privileges, and does not modify other skills or system-wide agent settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install burmese-audio-understanding - After installation, invoke the skill by name or use
/burmese-audio-understanding - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.2.2
- Added new Security Notes section detailing how audio data is handled and privacy considerations.
- Clarified that no data is stored locally after processing.
- Stated API key requirements and minimal permission guidance.
v1.2.1
- Added a "Required Environment Variables" section clarifying that `GEMINI_API_KEY` must be set for transcription services.
- Usage instructions and features remain unchanged.
v1.2.0
- Updated to use Gemini 3.1 Flash Preview for higher transcription accuracy.
- Switched to the official @google/genai SDK for secure file handling.
- Removed reliance on shell commands and ffmpeg; all audio processing is managed through the SDK.
- Skill name and description updated to reflect improved model and capabilities.
v1.1.0
- Added support for transcribing Burmese audio directly to text using Gemini 2.0 Flash.
- New script: `transcribe-direct.js` for audio file transcription.
- Now requires setting the `GEMINI_API_KEY` environment variable.
- Updated prerequisites: ffmpeg must be installed, and dependency on `@google/generative-ai` via npm.
Metadata
Frequently Asked Questions
What is Burmese Audio Understanding?
High-accuracy Burmese audio transcription using Gemini 3.1 Flash Preview. It is an AI Agent Skill for Claude Code / OpenClaw, with 182 downloads so far.
How do I install Burmese Audio Understanding?
Run "/install burmese-audio-understanding" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Burmese Audio Understanding free?
Yes, Burmese Audio Understanding is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Burmese Audio Understanding support?
Burmese Audio Understanding is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Burmese Audio Understanding?
It is built and maintained by thelapyae (@thelapyae); the current version is v1.2.2.
More Skills