songsee

Generate spectrograms and multi-panel audio feature visualizations from input files.

Updated Apr 5, 2026
One-click install
npx skills add https://github.com/dingdyan/openclaw-workspace-v2 --skill songsee-dingdyan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: songsee
Source: https://github.com/dingdyan/openclaw-workspace-v2/tree/main/skills/songsee
Command: npx skills add https://github.com/dingdyan/openclaw-workspace-v2 --skill songsee-dingdyan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Songsee helps you quickly visualize what is happening inside an audio file so you can analyze timbre, rhythm, and audio characteristics without manually running complex DSP pipelines.

Core Features & Use Cases

  • Spectrogram generation: Create clear time–frequency representations for audio inspection and presentation.
  • Feature-panel visualization: Render multiple audio feature views (e.g., MFCC, chroma, loudness, tempogram, flux) together for comparative analysis.
  • Targeted time slicing: Focus on a specific time window to inspect events like intros, beats, transitions, or anomalies.

Example use case: You have a song segment where vocals change abruptly and need to compare chroma, loudness, and MFCC behavior across a short time range; generate a multi-panel output for the full track or a focused slice.

Quick Start

Generate a spectrogram for your audio file by running: songsee track.mp3 -o output.png

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 audio file for visual analysis?

To generate an audio spectrogram, you can use the songsee CLI to decode your input file and render a time-frequency visual output by running a command like songsee track.mp3 -o output.png.

Can I visualize multiple audio features like MFCC and chroma in one view?

Yes, you can visualize multiple DSP-derived audio features together by rendering multi-panel outputs that display MFCC, chroma, loudness, and tempogram views side-by-side for comparative analysis.

How do I inspect a specific time window in an audio spectrogram?

You can inspect a specific time window in an audio spectrogram by using targeted time slicing to focus on exact segments, allowing you to analyze events like intros, beats, or transitions.

Do I need ffmpeg to analyze MP3 files for audio feature visualization?

You need ffmpeg for audio feature visualization when analyzing MP3 files that require decoding, although the CLI also supports native WAV file processing without additional dependencies.

What is the best way to compare loudness and MFCC behavior across a short audio segment?

The best way to compare loudness and MFCC behavior is by generating a multi-panel audio feature visualization that renders these DSP-derived characteristics together for a targeted time slice.

Can I configure the frequency range when creating spectrograms for music information retrieval?

Yes, you can configure the frequency range and windowing parameters when creating spectrograms for music information retrieval to customize the visual output for your specific analysis needs.