Can I customize the look of the generated spectrograms?

Yes, the skill supports multiple color palettes (magma, inferno, viridis, etc.) and allows you to define custom widths and heights for the output image.

What types of audio features can Songsee visualize?

You can visualize a wide range of features including Mel-frequency spectrograms, Chromagrams, MFCCs, loudness, tempograms, and self-similarity matrices.

What audio formats does the Songsee skill support?

It supports WAV and MP3 natively. Other formats can be processed if FFmpeg is installed and available on your system path.

How do I install the underlying tool for this skill?

The skill requires the songsee CLI, which can be installed via Homebrew using: brew install steipete/tap/songsee.

Can I visualize a specific part of an audio file?

Yes, you can use the --start and --duration flags to define a specific time slice for visualization.

Songsee Audio Visualization

Name: Songsee Audio Visualization
Author: ishuru

byishuru

0•

数据科学与机器学习

Generates high-quality audio spectrograms and multi-panel feature visualizations using the songsee CLI.

Songsee Audio Visualization is a specialized Claude Code skill designed for audio analysis and signal processing tasks. It enables Claude to programmatically generate detailed spectrograms and complex feature grids—including Mel scales, chromagrams, MFCCs, and loudness profiles—directly from audio files. Whether you are debugging audio algorithms, preparing datasets for machine learning, or creating visual assets for music documentation, this skill provides the necessary tools to inspect sound patterns with precision through customizable palettes, time-slicing, and high-resolution exports.

主要功能

01Native support for common formats with FFmpeg integration for extended compatibility.

02Generate detailed spectrograms and feature-panel visualizations from MP3/WAV files.

03Customizable visual styles with palettes like Magma, Inferno, Viridis, and Classic.

04Extract advanced audio features including Mel, Chroma, HPSS, MFCC, and Tempograms.

05Supports precise time-slicing for specific audio segments using start and duration flags.

060 GitHub stars

使用场景

01Generating visual documentation for audio processing pipelines and signal debugging.

02Visualizing audio features for machine learning preprocessing and data inspection.

03Analyzing frequency distributions and harmonic content in music production.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ishuru/ishuru songsee

For use in Claude.ai and ChatGPT