AI Blocks › Build and ship software
Signal feature extractor
Give it signals. Get time and frequency features ready for analysis or machine learning.
Signal feature extractor is a free AI skill for the job "Build and ship software". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · BSD 2 Clause licence · Made by MathWorks · Thinking: low
Made by MathWorks. See the original on GitHub; we copied it as it is, with its licence.
What it does
A table of time and frequency features for each frame of your signal, ready for analysis or a machine learning model. You get the right feature tool set up with real parameters only. It first checks whether frequency content shifts over time. Then it picks one of three MATLAB extractors: time, frequency or time-frequency. It sets the sample rate, frame size and frame rate or overlap, turns on the features you name, and runs the extraction. Per-frame results stay as they are unless your model needs averages. A GPU path is optional. Audio features like MFCC are left to other tools. It needs MATLAB with the Signal Processing Toolbox.
You give it
- A one-dimensional signal
- Sample rate and frame size
- Features you want (optional)
- MATLAB with Signal Processing Toolbox
You get
- A feature table or matrix per frame
- A configured extractor
- MATLAB code you can rerun
- An optional GPU version
Use it when
- You have vibration data and want features to train a classifier.
- You need RMS, crest factor or spectral values for every window.
- You want a feature table to feed into a model.
- You are unsure which feature extractor suits your signal.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can run MATLAB with Signal Processing Toolbox.