elumenotion/guideants/audiocpp-asr

Claude Code and agent skill by elumenotion/guideants

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Transcription beyond the GuideAnts wrapper contract: language hints and workspace files via the GPU host raw audiocpp gateway (/asr + /files) from a PC sandbox. Use when a transcription needs a language hint or the built-in ASR tool is too narrow.

Install

npx vskill@latest install elumenotion/guideants/audiocpp-asr

SKILL.md

audio.cpp extended ASR (experimental)

Paths — fixed layout, do not probe or re-derive. The sandbox CWD is the notebook's output directory. This skill's scripts live under Skills/audiocpp-asr/scripts/ relative to it. Write deliverables with bare filenames; never prefix with Output/.

Product ASR is multipart upload with no language control. This skill stages the file (/files) then calls raw /asr/v1/audio/transcriptions on the GPU host.

Environment (required for PC → the GPU host)

text
AUDIOCPP_SKILL_BASE_URL=http://<gpu-host-lan-ip>:8112/audiocpp-skill
AUDIOCPP_SKILL_TOKEN=<same as the GPU host GA_AUDIOCPP_SKILL_TOKEN>

An ASR model must already be loaded on the GPU host (GuideAnts Settings → Local models / API lifecycle).

Preflight

bash
python3 Skills/audiocpp-asr-extended/scripts/preflight.py --for asr-extended

Trust its verdict. With the gateway env set, it checks GPU host ASR via the gateway (not sandbox loopback).

Transcribe

bash
python3 Skills/audiocpp-asr-extended/scripts/engine_tool.py transcribe \
  uploads/clip.wav [--language de]

The script stages the file on the GPU host and returns JSON text. Engine model id is always qwen3-asr. Prefer WAV; other formats may need ffmpeg conversion first.

Sideload (advanced)

Fetching a non-catalog qwen3-family snapshot downloads onto the GPU host under /models-local/skill/asr/…. Loading it into the product ASR wrapper still requires GPU host-side /asr/admin/load (and replaces the user’s loaded model) — ask before doing that; prefer Settings / API lifecycle for durable loads.

bash
python3 Skills/audiocpp-asr-extended/scripts/fetch_model.py <hf-repo-id> \
  --dest /models-local/asr/<DirName>

Related

Speaker-labeled transcripts: audiocpp-diarize.

Reporting

End by telling the user what worked and what was blocked, quoting preflight evidence.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

See the full security report

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

See evaluation results

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/elumenotion/guideants/audiocpp-asr/badge)](https://verified-skill.com/skills/elumenotion/guideants/audiocpp-asr)

Something wrong with this skill or its page? Tell us.