🔒 Verified source: github.com/openai⏳ Pending test
📄🧭🔗📚🧭🗺️🔗
🧩 Teach this AI new skills
Unlock 8 new powers for your AI agent
Whisper has no skill store we can verify. 7 popular cross-AI skills for agents like Claude Code and Cursor, mapped to what your AI agent can do after you install them · tap a power
Pre-check: open Task Manager → Performance to see RAM and GPU. Then run this in Command Prompt to see GPU name, driver version and VRAM. Also check free disk space.
nvidia-smi
Do you see your GPU name and "Driver Version"?
Install Python 3.11 from python.org (tick "Add python.exe to PATH"). Then install ffmpeg (winget ships with Windows 10/11).
winget install --id Gyan.FFmpeg -e
Does ffmpeg -version work in a new terminal?
For NVIDIA GPUs: on pytorch.org/get-started pick Windows + Pip + CUDA and run the torch command shown there first (otherwise Whisper runs on CPU).
Transcribe an audio file. On a 4 GB GPU or CPU use the small model. The language is auto-detected; add e.g. --language English or --language Hindi to force it. .txt, .srt (subtitles) etc. are saved in the folder.
whisper audio.mp3 --model small
Did .txt and .srt files appear?
Pre-check: Task Manager → Performance → GPU shows "AMD Radeon …" and dedicated GPU memory (VRAM). Update the driver with AMD Software: Adrenalin (amd.com).
Can you see the GPU name and VRAM?
Install Python 3.11 from python.org (tick "Add python.exe to PATH"). Then install ffmpeg (winget ships with Windows 10/11).
winget install --id Gyan.FFmpeg -e
Does ffmpeg -version work in a new terminal?
On Windows there is no simple PyTorch path for AMD GPUs — run Whisper on CPU (small model), or see the faster-whisper / whisper.cpp guides for faster CPU options.
Transcribe an audio file. On a 4 GB GPU or CPU use the small model. The language is auto-detected; add e.g. --language English or --language Hindi to force it. .txt, .srt (subtitles) etc. are saved in the folder.
whisper audio.mp3 --model small
Did .txt and .srt files appear?
Pre-check: Apple menu → About This Mac. Note the Chip (Apple M-series or Intel), Memory (RAM) and macOS version.
Noted chip, memory and macOS version?
Install Python 3.11 and ffmpeg with Homebrew (brew.sh).
Transcribe an audio file. On a 4 GB GPU or CPU use the small model. The language is auto-detected; add e.g. --language English or --language Hindi to force it. .txt, .srt (subtitles) etc. are saved in the folder.
Transcribe an audio file. On a 4 GB GPU or CPU use the small model. The language is auto-detected; add e.g. --language English or --language Hindi to force it. .txt, .srt (subtitles) etc. are saved in the folder.
whisper audio.mp3 --model small
Did .txt and .srt files appear?
Pre-check: see your RAM (Task Manager / About This Mac / free -h). Small tools run without a GPU — just slower.
Know how much RAM you have?
Install as in your OS tab (skip the GPU torch step). On CPU use tiny/base/small — medium/turbo will be very slow.
whisper audio.mp3 --model base --language Hindi
Text file created?
Sizes are approximate and can change on the official page.
Errors & fixes
FileNotFoundError: [WinError 2] / "ffmpeg not found"
ffmpeg is missing or not on PATH. Install it and open a new terminal.
UserWarning: FP16 is not supported on CPU; using FP32 instead
Not an error — normal on CPU.
'whisper' is not recognized
The venv is not active. Run venv\Scripts\activate, or use python -m whisper audio.mp3.
Error building tiktoken / "No module named setuptools_rust"
Use Python 3.11; if it persists, pip install setuptools-rust (README) or install Rust.
"CUDA out of memory" / "torch.cuda.OutOfMemoryError"
Your GPU ran out of VRAM. Pick a smaller model or image size and close other apps.
"python is not recognized" / "command not found: python"
Python is missing or not on PATH. Install from python.org and tick "Add python.exe to PATH" on the first installer screen, then open a new terminal. On Mac/Linux use python3.
Popular AI skills that work with tools like Whisper 7 skills
Whisper has no skill store we can verify. These cross-AI skills (MCP servers and Agent Skills) work in most AI agents — each card shows one example install; see all AIs for yours.
MCPPlugins
🛡 Stay safe
Read first
Official first — from the AI maker or the service’s own GitHub org / domain.
Read SKILL.md and any scripts before installing. Skills can run code with your permissions.
Never paste base64 or “curl … | sh” download-and-run lines from a skill.
Stars, license and last update come from GitHub (17 Nov 2026). Commands are copied from official docs and are not tested on a real machine yet. Access tags are our summary — always check the source. Skills for every AI →