transcribe-audio-local

75 / 100 · Review before installing

Check before installing. Some of what it can do is expected: a coding skill usually runs build and test commands.

By Desko77/claude-code-skills-1c (GitHub) · coding · 95 stars · scanned at commit af48dc0

Локальная транскрибация аудиофайлов без отправки в облако. Используй когда пользователь просит транскрибировать запись, расшифровать аудио, сделать конспект встречи, преобразовать речь в текст. Только для аудио (m4a/mp3/wav/ogg/flac/aac/wma/opus). Движок: faster-whisper CUDA + опц. диаризация sherpa-onnx GPU (CUDA, RTF ~0.24). Поддерживает разделение по спикерам.

Works with any agent that supports SKILL.md skills.

What this skill can do

Downloads and installs other software

The author does not mention this.

Found in 7 places.

skills/transcribe-audio-local/README.md, line 138Linux: `sudo apt install ffmpeg`.
skills/transcribe-audio-local/README.md, line 152Установщик пробует `pip install sherpa-onnx -f https://k2-fsa.github.io/sherpa/onnx/cuda.html`. Если не вышло:
skills/transcribe-audio-local/README.md, line 157CPU-режим тоже работает: `pip install sherpa-onnx` без extra-index.

Show 4 more

skills/transcribe-audio-local/scripts/setup.py, line 137info(" - Linux: см. инструкцию вашего дистрибутива (apt install nvidia-driver-550 и т.п.)")
skills/transcribe-audio-local/scripts/setup.py, line 212raise RuntimeError(f"pip install упал на: {packages}")
skills/transcribe-audio-local/scripts/setup.py, line 248info(f" {py} -m pip install sherpa-onnx")
skills/transcribe-audio-local/scripts/setup.py, line 251info(f" {py} -m pip install <путь-к-wheel.whl>")

Runs commands on your computer

The author does not mention this. Expected for a coding skill.

Found in 18 places.

skills/transcribe-audio-local/SKILL.md, line 20python scripts/setup.py
skills/transcribe-audio-local/SKILL.md, line 37python ~/.claude/skills/transcribe-audio-local/scripts/setup.py
skills/transcribe-audio-local/SKILL.md, line 55python ~/.claude/skills/transcribe-audio-local/scripts/verify.py --full

Show 15 more

skills/transcribe-audio-local/SKILL.md, line 107PYTHONUNBUFFERED=1 PYTHONIOENCODING=utf-8 \
skills/transcribe-audio-local/scripts/diarize_sherpa.py, line 76subprocess.run(cmd, check=True, capture_output=True)
skills/transcribe-audio-local/scripts/setup.py, line 98return subprocess.run(cmd, **kwargs).returncode
skills/transcribe-audio-local/scripts/setup.py, line 113result = subprocess.run(
skills/transcribe-audio-local/scripts/setup.py, line 119except (subprocess.SubprocessError, OSError):
skills/transcribe-audio-local/scripts/transcribe.py, line 278def spawn_worker(mode: str, env_extra: dict[str, str], cli: list[str], python_exe: Path) -> subprocess.Popen:
skills/transcribe-audio-local/scripts/transcribe.py, line 284return subprocess.Popen(cmd, env=env)
skills/transcribe-audio-local/scripts/transcribe.py, line 287def spawn_sherpa_diarize(env_extra: dict[str, str], cli: list[str]) -> subprocess.Popen:
skills/transcribe-audio-local/scripts/transcribe.py, line 299return subprocess.Popen(cmd, env=env)
skills/transcribe-audio-local/scripts/verify.py, line 60result = subprocess.run(
skills/transcribe-audio-local/scripts/verify.py, line 69except (subprocess.SubprocessError, OSError) as e:
skills/transcribe-audio-local/scripts/verify.py, line 92result = subprocess.run(
skills/transcribe-audio-local/scripts/verify.py, line 106except subprocess.TimeoutExpired:
skills/transcribe-audio-local/scripts/verify.py, line 128r = subprocess.run(
skills/transcribe-audio-local/scripts/verify.py, line 138except (subprocess.SubprocessError, OSError) as e:

Connects to the internet

The author does not mention this.

Found in 2 places.

skills/transcribe-audio-local/scripts/setup.py, line 33import urllib.request
skills/transcribe-audio-local/scripts/setup.py, line 273with urllib.request.urlopen(url) as resp, open(tmp, "wb") as f:

Safer alternatives

Low-risk skills in the same category (coding) that look related. Check that one does what you need.

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'transcribe-audio-local' ] || [ -L ~/.agents/skills/'transcribe-audio-local' ]; then echo 'A skill folder named transcribe-audio-local already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/Desko77/claude-code-skills-1c' "$d" && git -C "$d" checkout --quiet 'af48dc0c29f1dbe68019a96fce30c2735b036aca' -- 'skills/transcribe-audio-local' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/transcribe-audio-local' ~/.agents/skills/'transcribe-audio-local' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'transcribe-audio-local' ~/.claude/skills/'transcribe-audio-local' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for Desko77/claude-code-skills-1c shows the score of its lowest-scoring skill.

SkillTrust badge for Desko77/claude-code-skills-1c

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/Desko77/claude-code-skills-1c.svg)](https://zalavijaysinh.github.io/skilltrust/repos/Desko77/claude-code-skills-1c/)

Score history

Date	Version	Score
2026-11-07	`af48dc0`	75