songsee

Generate spectrograms and audio feature visualizations from audio files via CLI.

1|Updated Mar 12, 2026
One-click install
npx skills add https://github.com/kaminocorp/hermes-alpha-hunter --skill songsee-kaminocorp
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: songsee
Source: https://github.com/kaminocorp/hermes-alpha-hunter/tree/main/skills/media/songsee
Command: npx skills add https://github.com/kaminocorp/hermes-alpha-hunter --skill songsee-kaminocorp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Analyzing audio content typically requires opening a DAW or writing custom signal-processing code. This Skill turns any audio file into visual representations—spectrograms, mel scales, chroma, MFCC, tempograms—so you can inspect frequency content, tempo, and structure at a glance. ## Core Features & Use Cases - Multi-Panel Visualizations: Render up to nine visualization types (spectrogram, mel, chroma, hpss, self-similarity, loudness, tempogram, MFCC, spectral flux) as a grid in a single image. - Flexible Input and Output: Accepts WAV and MP3 natively (other formats via ffmpeg), reads from stdin, and exports PNG or JPG with custom dimensions, color palettes, and time slices. - Use Case: A music producer debugging a synthesis pipeline can generate a mel spectrogram and spectral flux plot of the output, then feed the image to a vision model to verify the expected harmonic structure. ## Quick Start Generate a spectrogram from the file track.mp3 and save it as a PNG image for inspection.

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?▼

Run songsee with the audio file as an argument, for example songsee track.mp3, and it renders a spectrogram image. Use -o to set the output path and --format to choose png or jpg.

What audio visualization types does songsee support?▼

songsee supports nine visualization types: spectrogram, mel, chroma, hpss, self-similarity, loudness, tempogram, MFCC, and spectral flux. Pass them comma-separated via --viz to render multiple panels as a grid in one image.

Does songsee support audio formats other than WAV and MP3?▼

WAV and MP3 are decoded natively by songsee. Other formats such as FLAC or AAC require ffmpeg to be installed on the system for decoding.

How do I visualize only part of an audio file?▼

Use the --start and --duration flags to select a time slice, for example songsee track.mp3 --start 12.5 --duration 8 renders an 8-second segment starting at 12.5 seconds.

Can songsee read audio from stdin?▼

Yes, pipe audio data with a dash as the input, such as cat track.mp3 | songsee - --format png -o out.png. This is useful for chaining with other command-line audio tools.