SkillAgentSearch skills...

songsee

Visualize audio as mel spectrograms, chromagrams, MFCC (librosa)

Install / Use

npx skills add taracodlabs/aiden --skill songsee

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

85/100

Category

Other

Supported Platforms

Universal

Our assessment of songsee

songsee scores 85/100 on our quality scale, 221st of 331 Other skills we index.

Its SKILL.md is 4.9 KB long, well organised into 11 sections with 6 code examples: a solid amount of guidance for an agent.

It has 843 GitHub stars, a meaningful sign that others use it.

Substance
26/30
Structure
20/20
Description
12/15
Adoption
12/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated about 2 months ago, so songsee is actively maintained.
  • Our last check on 2026-11-05 found the source still online.
  • It is released under AGPL-3.0, a copyleft license: you can use it, but modified versions you distribute must carry the same license.
  • Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-10-06. Automated pattern scan on 2026-10-06. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

songsee compared with similar skills

All 4 of these similar skills score higher than songsee; compare them before choosing.

SkillScoreStarsUpdatedFormat
songsee (this skill)by taracodlabs858432mo agoSKILL.md
Agent-Reachby Panniantong100112.7k9d agoCLAUDE.md
headroomby headroomlabs-ai10079.9ktodayCLAUDE.md
Scraplingby D4Vinci100103.9ktodayMCP Server
crawl4aiby unclecode10087.5ktodayMCP Server

Frequently asked questions

How do I install songsee?
Run npx skills add taracodlabs/aiden --skill songsee. The install tabs above show the steps for each supported agent.
Which AI agents does songsee work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is songsee safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is AGPL-3.0-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is songsee still maintained?
The repository was last updated about 2 months ago, so songsee is actively maintained.

name: songsee description: Visualize audio as mel spectrograms, chromagrams, MFCC (librosa) category: media version: 1.0.0 origin: aiden license: Apache-2.0 tags: audio, spectrogram, mel, chroma, mfcc, librosa, visualization, music, sound-analysis

Audio Visualization with Spectrograms

Visualize audio files as mel spectrograms, chromagrams, and MFCC feature plots using the librosa Python library. Useful for music analysis, speech processing, and audio debugging.

When to Use

  • User wants to visualize what an audio file "looks like"
  • User wants to analyze the frequency content of a recording
  • User wants to compare two audio files visually
  • User wants to understand musical key or chroma content
  • User wants to extract MFCC features for a machine learning task

How to Use

1. Install dependencies

pip install librosa matplotlib soundfile

2. Generate a mel spectrogram

import librosa
import librosa.display
import matplotlib.pyplot as plt
import numpy as np

def mel_spectrogram(audio_path, output="mel_spec.png"):
  y, sr = librosa.load(audio_path, sr=None)
  S     = librosa.feature.melspectrogram(y=y, sr=sr, n_mels=128, fmax=8000)
  S_db  = librosa.power_to_db(S, ref=np.max)

  fig, ax = plt.subplots(figsize=(12, 4), facecolor="#0d1117")
  ax.set_facecolor("#0d1117")
  img = librosa.display.specshow(S_db, sr=sr, x_axis="time", y_axis="mel", fmax=8000, ax=ax, cmap="magma")
  fig.colorbar(img, ax=ax, format="%+2.0f dB", label="dB")
  ax.set_title(f"Mel Spectrogram — {audio_path}", color="white")
  ax.tick_params(colors="white")
  ax.xaxis.label.set_color("white")
  ax.yaxis.label.set_color("white")
  plt.tight_layout()
  plt.savefig(output, dpi=150, bbox_inches="tight")
  plt.close()
  print(f"Saved: {output}")

mel_spectrogram("song.mp3")

3. Generate a chromagram (musical key content)

import librosa, librosa.display, matplotlib.pyplot as plt

def chromagram(audio_path, output="chroma.png"):
  y, sr   = librosa.load(audio_path, sr=None)
  chroma  = librosa.feature.chroma_cqt(y=y, sr=sr)

  fig, ax = plt.subplots(figsize=(12, 4), facecolor="#0d1117")
  ax.set_facecolor("#0d1117")
  img = librosa.display.specshow(chroma, y_axis="chroma", x_axis="time", ax=ax, cmap="coolwarm")
  fig.colorbar(img, ax=ax)
  ax.set_title("Chromagram", color="white")
  ax.tick_params(colors="white")
  plt.tight_layout()
  plt.savefig(output, dpi=150, bbox_inches="tight")
  plt.close()
  print(f"Saved: {output}")

chromagram("song.mp3")

4. Generate MFCC features

import librosa, librosa.display, matplotlib.pyplot as plt
import numpy as np

def mfcc_plot(audio_path, n_mfcc=20, output="mfcc.png"):
  y, sr = librosa.load(audio_path, sr=None)
  mfccs = librosa.feature.mfcc(y=y, sr=sr, n_mfcc=n_mfcc)

  fig, ax = plt.subplots(figsize=(12, 4), facecolor="#0d1117")
  ax.set_facecolor("#0d1117")
  img = librosa.display.specshow(mfccs, x_axis="time", ax=ax, cmap="viridis")
  fig.colorbar(img, ax=ax)
  ax.set_title(f"MFCC ({n_mfcc} coefficients)", color="white")
  ax.tick_params(colors="white")
  plt.tight_layout()
  plt.savefig(output, dpi=150, bbox_inches="tight")
  plt.close()
  print(f"Saved: {output}")

mfcc_plot("speech.wav", n_mfcc=13)

5. Generate all three plots at once

def analyze_audio(audio_path):
  base = audio_path.rsplit(".", 1)[0]
  mel_spectrogram(audio_path, output=f"{base}_mel.png")
  chromagram(audio_path,      output=f"{base}_chroma.png")
  mfcc_plot(audio_path,       output=f"{base}_mfcc.png")
  print(f"Analysis complete: 3 PNG files saved for {audio_path}")

analyze_audio("recording.wav")

6. Get basic audio statistics

import librosa, numpy as np

y, sr     = librosa.load("audio.mp3", sr=None)
duration  = librosa.get_duration(y=y, sr=sr)
tempo, _  = librosa.beat.beat_track(y=y, sr=sr)
rms       = np.sqrt(np.mean(y**2))

print(f"Duration:   {duration:.2f} seconds")
print(f"Sample rate:{sr} Hz")
print(f"Tempo:      {tempo:.1f} BPM")
print(f"RMS energy: {rms:.4f}")

Examples

"Show me what this audio recording looks like as a spectrogram" → Use step 2 to generate a mel spectrogram PNG. Open the saved file.

"What musical key is this song in? Visualize the chroma content" → Use step 3 to generate a chromagram — peaks in chroma rows indicate dominant pitch classes.

"Generate MFCC features from this speech recording for my ML model" → Use step 4 to plot MFCCs, then extract the mfccs array for downstream ML use.

Cautions

  • librosa loads audio in float32 mono by default — stereo files are mixed down automatically
  • Large audio files (> 30 minutes) take significant time and memory to process — slice with offset and duration parameters if needed
  • librosa.load supports MP3, WAV, FLAC, OGG — ensure soundfile and audioread are installed for MP3 support
  • MFCC coefficients are sensitive to n_mfcc and sr — use consistent settings across all files in an ML dataset

Related Skills

View on GitHub
GitHub Stars843
CategoryOther
Updated2mo ago
Forks147

Languages

TypeScript

Trust signals

100/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

No cautions