songsee

What This Skill Does

The songsee skill provides a powerful command-line interface for audio analysis, enabling users to transform raw audio files into detailed visual representations. At its core, songsee leverages advanced digital signal processing to generate spectrograms and a wide variety of feature panels. Whether you are a music producer looking to visualize harmonic content, a researcher analyzing sound profiles, or a developer needing to inspect audio data, this tool offers precise control over visualization outputs. The skill handles common formats like WAV and MP3 natively, while also utilizing ffmpeg for extended compatibility, ensuring that almost any audio file can be processed.

Installation

To integrate this tool into your OpenClaw environment, execute the following command in your terminal:

clawhub install openclaw/skills/skills/steipete/songsee

Ensure that you have ffmpeg installed on your system if you intend to work with non-standard audio formats, as the tool relies on it for robust decoding beyond basic native support.

Use Cases

• Audio Archiving: Create high-quality visual thumbnails for large audio libraries to quickly identify tracks.
• Music Production Analysis: Use chroma and tempogram visualizations to understand the harmonic and rhythmic structure of a track.
• Signal Inspection: Isolate specific time intervals using the --start and --duration flags to examine transient responses or noise floors.
• Automated Reporting: Pipe audio streams directly into songsee to generate automated performance reports during batch processing.

Example Prompts

1. "Analyze my file 'recording.wav' and generate a grid containing the spectrogram, chroma, and mfcc visualizations."
2. "Extract a 10-second segment from 'drums.mp3' starting at the 30-second mark and save it as a high-quality png file using the magma color palette."
3. "Generate a multi-panel visual for 'song.mp3' featuring the loudness, flux, and tempogram to help me identify the peak energy sections."

Tips & Limitations

• Efficiency: When rendering large grids with many visualizations, ensure sufficient system memory, as plotting high-resolution graphs can be resource-intensive.
• FFmpeg: Always check that your ffmpeg path is correctly configured in your environment variables to avoid decoding errors for proprietary formats.
• Customization: Use the --style flag to adjust visual themes; 'magma' or 'inferno' are often preferred for highlighting peak intensities in spectrograms compared to 'classic' or 'gray'.
• Precision: If you are working with small windows, use the --window and --hop flags to tune the FFT sensitivity to your specific audio frequency requirements.