songsee

Generate spectrograms and feature panels from audio via the songsee CLI.

Updated Mar 23, 2026

Search Tags:#cli #visualization #audio #spectrogram #feature-panels #songsee

Ready to run

npx skills add https://github.com/Kastalien-Research/kastalien-factory --skill songsee-kastalien-research

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: songsee
Source: https://github.com/Kastalien-Research/kastalien-factory/tree/main/lettabot/.skills/songsee
Command: npx skills add https://github.com/Kastalien-Research/kastalien-factory --skill songsee-kastalien-research

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Generate quick, interpretable visualizations of audio by turning tracks into spectrograms and feature panels. This enables analysts and creators to quickly inspect spectral content, compare segments, and identify distinctive audio features without manual plotting.

Core Features & Use Cases

Spectrogram generation from WAV/MP3 native decoding with optional format support via ffmpeg.
Feature-panel visualizations including mel, chroma, hpss, self-similarity, loudness, tempogram, and MFCCs.
Use Case: rapidly visualize a playlist of tracks for quality checks or podcast analysis.

Quick Start

Run songsee track.mp3 to generate the default spectrogram and feature-panel outputs.

Frequently Asked Questions about songsee

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate a spectrogram from an MP3 file?▼

To generate a spectrogram from an MP3 file, you can use a CLI tool to decode the audio and output visual spectrograms. The songsee Skill natively decodes WAV and MP3 files to produce interpretable spectrograms and feature panels for audio analysis.

What audio feature visualizations can I create for podcast analysis?▼

You can generate feature-panel visualizations including mel, chroma, hpss, self-similarity, loudness, tempogram, and MFCCs. These feature panels allow you to inspect spectral content and compare segments for podcast or track analysis.

Do I need ffmpeg to visualize spectrograms for audio formats other than WAV or MP3?▼

You need ffmpeg to process audio formats beyond native WAV and MP3 decoding. The tool supports native WAV and MP3 decoding, but requires ffmpeg installed to decode and visualize additional audio formats for spectrogram generation.

Can I customize the frequency range and size of generated audio spectrograms?▼

You can customize the frequency range and size of generated audio spectrograms. The tool provides options to customize visualization panels, output size, and frequency range to suit specific audio analysis needs.

What is the best way to compare spectral content across multiple audio segments?▼

The best way to compare spectral content is by generating spectrograms and feature panels like self-similarity and mel. This allows analysts to visually inspect and compare distinctive audio features across multiple segments without manual plotting.