القائمة
Advanced audio analysis and feature extraction using librosa with optional Vamp plugin support via sonic-annotator
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | daw-master:audio-analyzer |
| version | 1.0.0 |
| description | Advanced audio analysis and feature extraction using librosa with optional Vamp plugin support via sonic-annotator |
| author | Hermes Agent |
| license | MIT |
| metadata | {"hermes":{"tags":["audio","analysis","feature-extraction","librosa","vamp","statistics"],"provides":["analyze","probe","extract_batch"],"requires":["python","librosa","soundfile","numpy"],"optional":["sonic-annotator"]}} |
Comprehensive audio analysis skill for extracting musical features, spectral characteristics, and metadata from audio files. Integrates with the daw-master meta-skill pipeline.
| Operation | Description | Parameters |
|---|---|---|
analyze | Extract comprehensive feature set from a single audio file | file, features (list, optional subset) |
probe | Basic file metadata (duration, sample rate, channels, format) | file |
extract_batch | Recursively analyze directory, write results to CSV/JSON | directory, pattern, output_format, output_file |
Required (Python):
pip install librosa soundfile numpy
Optional (Vamp plugins):
sudo apt install sonic-annotator
# Install Vamp plugin packs:
# qm-vamp-plugins: tempo tracking, chroma, onset detection
# vamp-example-plugins: basic features
from skills.daw_master.audio_analyzer import analyze, probe, extract_batch
# Single file analysis
features = analyze("input.wav")
print(f"Tempo: {features['tempo']} BPM")
print(f"Key: {features['key']}")
# Batch processing
extract_batch(
directory="./samples",
pattern="*.wav",
output_format="csv",
output_file="features.csv"
)
analyze() returns a dictionary containing:
{
"duration": 180.5,
"sample_rate": 44100,
"channels": 2,
"tempo": 128.0,
"tempo_confidence": 0.85,
"key": "C major",
"key_confidence": 0.72,
"loudness": -12.3,
"spectral_centroid": 2450.0,
"spectral_rolloff": 5800.0,
"spectral_flatness": 0.15,
"mfcc": [mean_vector, std_vector],
"harmonic_ratio": 0.68,
"percussive_ratio": 0.32,
"onset_count": 240,
"zero_crossing_rate": 0.08
}
All audio-analyzer operations are pipeable:
# Chain with sox-engine for preprocessing
from skills.daw_master.sox_engine import trim, normalize
from skills.daw_master.audio_analyzer import analyze
# Normalize then analyze
normalize("raw.wav", "normalized.wav")
features = analyze("normalized.wav")
pattern="**/*.wav" for recursive directory scanningvamp_plugins key in outputUse when working with the Sleepy Circuits Hypno 2 video synthesizer/resampler — 2-channel video mixer/looper, shader engine, MIDI CC mapping, CV/modulation, sampler. Firmware v0.0.163+.
Hardware instrument skills — reference guides for each device in the music studio setup.
GLSL fragment shader techniques adapted for the Hypno 2 video synthesizer — single-pass generative visuals and video effects optimized for Raspberry Pi 5, 5-uniform limit, CC-mapped parameters
Korg KAOSS DJ — USB DJ controller with X-Y touchpad, Serato DJ Intro integration, KAOSS effects, EQ, crossfader, and MIDI on Linux
Korg KAOSSILATOR dynamic phrase synthesizer (2007 model) — X-Y touchpad, 100 programs, gate arpeggiator, phrase loop recording, USB MIDI on Linux
Use when working with the Korg kaossilator 2S dynamic phrase synthesizer — X-Y touchpad, 150 programs, 50 arp patterns, loop recording, audio player, master recorder, microSD storage, and USB MIDI on Linux.