Groq Voice Transcribe

Transcribe audio files via Groq's OpenAI-compatible speech-to-text API. Use when the user sends voice messages or audio files and you need fast cloud speech-...

Security Scan

The skill appears to do what it claims (upload audio to Groq's transcription API) but the registry metadata omits the required GROQ_API_KEY and the SKILL.md instructs the assistant to write your API key into ~/.openclaw/openclaw.json — these inconsistencies and the lack of provenance warrant caution.

Purpose & Capability

Name/description align with the included script and purpose (cloud transcription via Groq). However, the registry metadata lists no required environment variables or primary credential while the script and SKILL.md clearly require a GROQ_API_KEY — an incoherence between declared requirements and actual need.

Instruction Scope

SKILL.md and scripts are narrowly scoped to sending an audio file to Groq's transcription endpoint and saving the response. The SKILL.md additionally tells the assistant it can place the API key into ~/.openclaw/openclaw.json — an action that writes user configuration and involves handling a secret. This is explainable for setup but should be done explicitly and securely.

Install Mechanism

No install spec; the skill is instruction + a small shell script. Nothing is downloaded from external/unknown URLs or installed automatically, which keeps installation risk low.

Credentials

Functionally the skill needs one secret (GROQ_API_KEY) which is proportionate. The concern is that registry metadata did not declare any required env vars or primary credential despite the script depending on GROQ_API_KEY; this mismatch is an integrity problem and could lead to misconfiguration or accidental key disclosure if users follow unclear instructions.

Persistence & Privilege

No elevated privileges requested. always is false and the skill does not request persistent system-level changes beyond optionally writing the API key into the local OpenClaw config (which is limited in scope).

What to consider before installing

This skill's code does exactly what it says: it uploads a local audio file to https://api.groq.com/openai/v1/audio/transcriptions using a GROQ_API_KEY and writes the transcription to a file. Before installing or using it: - Treat the GROQ API key as a secret. Do not paste it into public chats. Prefer setting GROQ_API_KEY in your environment or adding it manually to ~/.openclaw/openclaw.json rather than letting the assistant edit config automatically. - The registry metadata did not list any required env var, but the script requires GROQ_API_KEY — that's an inconsistency. Confirm where you store the key and how it will be used. - The skill will send audio to Groq's cloud for transcription. Only use it for audio you are comfortable sending to an external service. - The package has no install step and the included script is small and readable; you can run the script manually to verify behavior before granting the assistant permission to configure anything. - The skill source and homepage are unknown. If you need stronger provenance, ask the publisher for a homepage or repository and prefer skills with clear authorship. If you decide to proceed: set GROQ_API_KEY locally (export GROQ_API_KEY=...) and run the provided script manually to test, or explicitly instruct the assistant how/where to store the key (and confirm it will not be shared elsewhere).

Fast speech-to-text for voice notes and audio files through Groq's OpenAI-compatible transcription endpoint. Use it when you want cloud transcription via Groq instead of running Whisper locally.

Best for:

• Telegram / Signal voice notes
• short audio clips
• Chinese, English, or mixed daily speech
• fast transcript generation for follow-up summarization or chat replies

What you need

You need a Groq API key. Groq often provides a free developer tier / trial credits for new users. Get one from:

• https://console.groq.com/
• Sign in, open API Keys, then create a key

Easiest setup in OpenClaw

If OpenClaw is already running and configured, you can simply ask your assistant:

• "Configure Groq Voice Transcribe for me"
• "Here is my Groq API key, set up Groq Voice Transcribe"

The assistant can place the key into ~/.openclaw/openclaw.json for you.

Manual setup

Set GROQ_API_KEY, or configure it in ~/.openclaw/openclaw.json under:

{
  "skills": {
    "entries": {
      "groq-voice-transcribe": {
        "apiKey": "GROQ_KEY_HERE"
      }
    }
  }
}

Quick start

{baseDir}/scripts/transcribe.sh /path/to/audio.ogg

Defaults:

• Model: whisper-large-v3-turbo
• Output: <input>.txt
• Format: plain text

Common examples

# Basic transcript
{baseDir}/scripts/transcribe.sh /path/to/audio.ogg

# Chinese voice message
{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --language zh --prompt "中文普通话，日常聊天"

# Save to a custom file
{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --out /tmp/transcript.txt

# Verbose JSON output
{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --json --out /tmp/transcript.json

Flags

• --model <name>: transcription model (default whisper-large-v3-turbo)
• --out <path>: output file path
• --language <code>: hint the spoken language, for example zh, en, ja
• --prompt <text>: optional context or spelling hint
• --json: write verbose JSON instead of plain text

Notes

• Audio is sent to Groq for transcription.
• This skill is meant for transcription, not text-to-speech.
• If language is omitted, Groq can usually auto-detect it, but passing --language zh often helps for Chinese voice notes.