transcribe-audio-local

transcribe-audio-local is a skill for Claude Code, Codex from Desko77/cursor-1c-skills. It costs 121 tokens per session (1,620 once invoked), scanned A, a copy of transcribe-audio-local, MIT.

A local tool that turns audio files such as MP3, WAV, or M4A into written text without sending the recording to an online service. It can optionally identify different speakers.

In plain words
What is it for?
Transcribe recordings, turn speech into text, and create meeting notes from audio files.
Why use it?
It keeps recordings private and avoids relying on cloud transcription services. It also provides a transcript for audio-only files.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: reads .claude/ paths; mentions Claude Code.

Good fit Transcribe recordings, turn speech into text, and create meeting notes from audio files.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/desko77/cursor-1c-skills/transcribe-audio-local
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add Desko77/cursor-1c-skills --skill transcribe-audio-local
Clone the repo
git clone --depth 1 https://github.com/Desko77/cursor-1c-skills

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for transcribe-audio-local

README.md
[![agentmods](https://agentmods.dev/badge/skills/desko77/cursor-1c-skills/transcribe-audio-local/github.svg)](https://agentmods.dev/skills/desko77/cursor-1c-skills/transcribe-audio-local)
Your own site
<a href="https://agentmods.dev/skills/desko77/cursor-1c-skills/transcribe-audio-local"><img src="https://agentmods.dev/badge/skills/desko77/cursor-1c-skills/transcribe-audio-local/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for transcribe-audio-local

Your own site · 80×15
<a href="https://agentmods.dev/skills/desko77/cursor-1c-skills/transcribe-audio-local"><img src="https://agentmods.dev/badge/skills/desko77/cursor-1c-skills/transcribe-audio-local.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 121 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,620 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium analysis-evasion · line 1
    Suspicious Unicode normalization or mixed-script content
    Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
How audits are shown
Origin 91% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00121 $0.01620
Opus 5 $0.00060 $0.00810
Sonnet 5 $0.00024 $0.00324
Haiku 4.5 $0.00012 $0.00162

Measured 7d ago against content hash dc477bd84c41, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

transcribe-audio-local scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.

The scan reads SKILL.md. This mod also ships 4 executable files (scripts/diarize_sherpa.py, scripts/setup.py, scripts/transcribe.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

91% identical to transcribe-audio-local — 45 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

skills/transcribe-audio-local/SKILL.md · 126 lines

How it starts

The opening of the file, as written. The whole thing — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/transcribe-audio-local - Локальная транскрибация аудио

Локальный движок: faster-whisper (CUDA) + опц. диаризация sherpa-onnx GPU (CUDA) с моделями pyannote-segmentation-3.0 + 3D-Speaker eres2net. Нет затрат, ничего не уходит наружу.

На RTX 5070 Ti Laptop: ~6-7 мин на 30 мин аудио (RTF ~0.24). Только аудио, видео не поддерживается.

Установка (один раз)

python scripts/setup.py

Подробнее в README.md. После установки venv-whisper и venv-sherpa создаются рядом со скилом, модели качаются в models/.

Для агента (Claude Code и подобные)

Workflow первого использования:

  1. Проверь, установлен ли скил - есть ли файл ~/.claude/skills/transcribe-audio-local/venv-whisper/Scripts/python.exe (на Linux: venv-whisper/bin/python). Если есть - переходи к шагу 4.

  2. Если не установлен - запусти setup. Установка занимает 10-20 минут (faster-whisper + ctranslate2-CUDA + sherpa-onnx + скачивание моделей). Дефолтный Bash-таймаут 2 минуты не подходит:

  • В Claude Code Bash tool используй run_in_background: true или timeout: 1500000 (25 мин).
  • В других runtime - аналогичный длинный таймаут.
python ~/.claude/skills/transcribe-audio-local/scripts/setup.py

Что setup сделает сам:

  • Создаст venv-whisper и venv-sherpa
  • Если ffmpeg нет в PATH - поставит static-ffmpeg (pip-пакет с бинарниками) в venv-whisper. Без прав админа.
  • Скачает модели диаризации
  • Проверит версию драйверов NVIDIA и предупредит, если они старее 525.x

Что setup НЕ может сделать сам (нужны действия пользователя):

  • Установить системные драйверы NVIDIA / CUDA Toolkit (нужны админ-права)
  • Поставить NVIDIA GPU в машину 😉

Если у машины нет GPU - setup упадет с FAIL. Уточни у пользователя, согласен ли на CPU-режим (в 10+ раз медленнее), и перезапусти с --allow-cpu.

  1. Проверь установку smoke-тестом (1-2 минуты, с реальной транскрипцией):
python ~/.claude/skills/transcribe-audio-local/scripts/verify.py --full

Без --full (5 сек) - только проверка наличия venv'ов и моделей, без реального прогона.

Read the full file on GitHub · 126 lines

Files

What ships with it

5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 7d ago First seen · 126 lines · 121 tokens per session scan A dc477bd84c41

Subscribe to this mod's changes

transcribe-audio-local is a skill published in the GitHub repository Desko77/cursor-1c-skills (57 stars, last pushed today), licensed MIT. It adds 121 tokens to every session and 1,620 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. It is 91% identical to transcribe-audio-local, differing in 45 lines, and is treated as a copy.