← Back to Skills Marketplace
Voice Transcribe
by
darinkishore
· GitHub ↗
· v1.0.1
6391
Downloads
13
Stars
41
Active Installs
2
Versions
Install in OpenClaw
/install voice-transcribe
Description
Transcribe audio files using OpenAI's gpt-4o-mini-transcribe model with vocabulary hints and text replacements. Requires uv (https://docs.astral.sh/uv/).
Usage Guidance
Install only if you trust the local transcribe script at the referenced path or can review it separately. Use a dedicated OpenAI API key, avoid sensitive or third-party voice recordings unless you have consent, and check where transcripts or cache files are stored and how to clear them.
Capability Analysis
Type: OpenClaw Skill
Name: voice-transcribe
Version: 1.0.1
The skill bundle describes an audio transcription tool using OpenAI's API. It instructs the agent to store an `OPENAI_API_KEY` locally for its intended use and to run a local script (`transcribe`). There is no evidence of data exfiltration, malicious execution, persistence mechanisms, prompt injection attempts, or obfuscation within the provided `_meta.json`, `SKILL.md`, `replacements.txt`, and `vocab.txt` files. All instructions and configurations are directly aligned with the stated purpose of transcribing audio.
Capability Assessment
Purpose & Capability
The transcription purpose is coherent, but the main capability depends on a local transcribe executable that is not included in the reviewed artifact, so its actual behavior cannot be verified from the package.
Instruction Scope
SKILL.md tells the agent to 'just run' transcription for voice memos, especially WhatsApp audio, without explicit confirmation or sensitivity screening before third-party processing.
Install Mechanism
The command points to /Users/darin/clawd/skills/voice-transcribe/transcribe and the setup stores credentials in that same personal path; there is no bundled script or install spec.
Credentials
Using OpenAI and an OPENAI_API_KEY is proportionate for transcription, but users should understand audio may be sent to OpenAI and API costs may be incurred.
Persistence & Privilege
The skill uses a local .env key, editable vocabulary/replacement files, and a stated cache by audio hash; these are purpose-aligned but cache location and cleanup are not documented.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install voice-transcribe - After installation, invoke the skill by name or use
/voice-transcribe - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
Added uv requirement to description
v1.0.0
Initial release: transcribe audio with gpt-4o-mini-transcribe
Metadata
Frequently Asked Questions
What is Voice Transcribe?
Transcribe audio files using OpenAI's gpt-4o-mini-transcribe model with vocabulary hints and text replacements. Requires uv (https://docs.astral.sh/uv/). It is an AI Agent Skill for Claude Code / OpenClaw, with 6391 downloads so far.
How do I install Voice Transcribe?
Run "/install voice-transcribe" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Voice Transcribe free?
Yes, Voice Transcribe is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Voice Transcribe support?
Voice Transcribe is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Voice Transcribe?
It is built and maintained by darinkishore (@darinkishore); the current version is v1.0.1.
More Skills