Local Whisper

This is OpenAI's Whisper running locally on your machine, which means you can transcribe audio without sending files to an API. It's genuinely offline after the initial model download, and you get five model sizes to pick from depending on whether you care more about speed or accuracy. The base model is 74MB and works fine for most stuff, but turbo at 809MB is probably the sweet spot if you have the disk space. It outputs plain text by default but supports timestamps and JSON if you need structured data. Requires ffmpeg and uses a uv-managed Python environment, so setup is straightforward enough. Good option if you're dealing with sensitive audio or just don't want to pay per minute for transcription.

skills.sh installs: 297
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add thinkfleetai/thinkfleet-engine --skill local-whisper --agent claude-code

Installs into .claude/skills of the current project.

Files

Local Whisper STT

Local speech-to-text using OpenAI's Whisper. Fully offline after initial model download.

Usage

# Basic
~/.thinkfleetbot/skills/local-whisper/scripts/local-whisper audio.wav

# Better model
~/.thinkfleetbot/skills/local-whisper/scripts/local-whisper audio.wav --model turbo

# With timestamps
~/.thinkfleetbot/skills/local-whisper/scripts/local-whisper audio.wav --timestamps --json

Models

Model	Size	Notes
`tiny`	39M	Fastest
`base`	74M	Default
`small`	244M	Good balance
`turbo`	809M	Best speed/quality
`large-v3`	1.5GB	Maximum accuracy

Options

--model/-m — Model size (default: base)
--language/-l — Language code (auto-detect if omitted)
--timestamps/-t — Include word timestamps
--json/-j — JSON output
--quiet/-q — Suppress progress

Setup

Uses uv-managed venv at .venv/. To reinstall:

cd ~/.thinkfleetbot/skills/local-whisper
uv venv .venv --python 3.12
uv pip install --python .venv/bin/python click openai-whisper torch --index-url https://download.pytorch.org/whl/cpu

SKILL.md on GitHub