Add video-to-spectrogram batch conversion utilities using existing STFT functions

[Copilot]

Implements standalone utilities for converting audio/video files to spectrogram images using the existing fourier_transformation and make_logscale functions from the Video Node, enabling batch dataset preparation for audio classification tasks.

Changes

Core Scripts

• simple_video_to_spectrogram.py - Minimal batch processor following ESC-50 CSV pattern with fourier_transformation, make_logscale, and plot_spectrogram functions
• video_to_spectrogram.py - Full CLI tool with single/batch modes, video support via ffmpeg audio extraction, configurable FFT bins and colormaps

Documentation & Examples

• VIDEO_TO_SPECTROGRAM_README.md - Technical reference with parameters, output structure, performance tips
• QUICKSTART_VIDEO_TO_SPECTROGRAM.md - Quick start guide with common use cases
• examples/video_to_spectrogram_example.py - Working examples for single file, batch, and ESC-50 processing
• Updated main README with feature section and usage examples

Testing

• tests/test_video_to_spectrogram.py - Integration tests for STFT, log scaling, and end-to-end spectrogram generation
• Updated requirements.txt with scipy and pandas

Usage

Batch processing with CSV metadata:

from simple_video_to_spectrogram import process_video_chunks_to_spectrograms

process_video_chunks_to_spectrograms(
    csv_path='metadata.csv',  # columns: filename, category
    audio_root='audio/',
    spectrogram_root='spectrograms/'
)

CLI for single files:

python video_to_spectrogram.py --mode single --input video.mp4 --output spec.jpg --binsize 2048

Spectrograms use identical STFT parameters as the Video Node (1024 bins, 50% overlap, Hanning window) with amplitude-to-decibel conversion and configurable colormaps.

Original prompt

utilise ça : def fourier_transformation(sig, frameSize, overlapFac=0.5, window=np.hanning):
win = window(frameSize)
hopSize = int(frameSize - np.floor(overlapFac * frameSize))

# zeros at beginning (thus center of 1st window should be for sample nr. 0)
samples = np.append(np.zeros(int(np.floor(frameSize/2.0))), sig)
# cols for windowing
cols = np.ceil( (len(samples) - frameSize) / float(hopSize)) + 1
# zeros at end (thus samples can be fully covered by frames)
samples = np.append(samples, np.zeros(frameSize))

frames = stride_tricks.as_strided(samples, shape=(int(cols), frameSize), strides=(samples.strides[0]*hopSize, samples.strides[0])).copy()
frames *= win

return np.fft.rfft(frames)   def make_logscale(spec, sr=44100, factor=20.):
timebins, freqbins = np.shape(spec)

scale = np.linspace(0, 1, freqbins) ** factor
scale *= (freqbins-1)/max(scale)
scale = np.unique(np.round(scale))

# create spectrogram with new freq bins
newspec = np.complex128(np.zeros([timebins, len(scale)]))
for i in range(0, len(scale)):
    if i == len(scale)-1:
        newspec[:,i] = np.sum(spec[:,int(scale[i]):], axis=1)
    else:
        newspec[:,i] = np.sum(spec[:,int(scale[i]):int(scale[i+1])], axis=1)

# list center freq of bins
allfreqs = np.abs(np.fft.fftfreq(freqbins*2, 1./sr)[:freqbins+1])
freqs = []
for i in range(0, len(scale)):
    if i == len(scale)-1:
        freqs += [np.mean(allfreqs[int(scale[i]):])]
    else:
        freqs += [np.mean(allfreqs[int(scale[i]):int(scale[i+1])])]

return newspec, freqs   import os

import pandas as pd
import scipy.io.wavfile as wav
import numpy as np
import matplotlib.pyplot as plt

Ta fonction plot_spectrogram

def plot_spectrogram(location, plotpath=None, binsize=2**10, colormap="jet"):
samplerate, samples = wav.read(location)
s = fourier_transformation(samples, binsize)
sshow, freq = make_logscale(s, factor=1.0, sr=samplerate)
ims = 20.*np.log10(np.abs(sshow)/10e-6) # amplitude to decibel

timebins, freqbins = np.shape(ims)
#print("timebins:", timebins, "freqbins:", freqbins)

plt.figure(figsize=(15, 7.5))
plt.imshow(np.transpose(ims), origin="lower", aspect="auto", cmap=colormap, interpolation="none")
xlocs = np.float32(np.linspace(0, timebins-1, 5))
plt.xticks(xlocs, ["%.02f" % l for l in ((xlocs*len(samples)/timebins)+(0.5*binsize))/samplerate])
ylocs = np.int16(np.round(np.linspace(0, freqbins-1, 10)))
plt.yticks(ylocs, ["%.02f" % freq[i] for i in ylocs])

if plotpath:
    plt.savefig(plotpath, bbox_inches="tight")
else:
    plt.show()
plt.clf()

return ims

Charger le CSV

esc50_df = pd.read_csv('/content/ESC-50-master/meta/esc50.csv')

Créer les dossiers

spectrogram_root = '/content/ESC-50-master/spectrogram'
os.makedirs(spectrogram_root, exist_ok=True)

for cat in esc50_df['category'].unique():
os.makedirs(os.path.join(spectrogram_root, cat), exist_ok=True)

Générer tous les spectrogrammes

for i, row in esc50_df.iterrows():
filename = row['filename']
category = row['category']
audio_path = os.path.join('/content/ESC-50-master/audio', filename)
save_path = os.path.join(spectrogram_root, category, filename.replace('.wav', '.jpg'))

try:
    plot_spectrogram(audio_path, plotpath=save_path)
except Exception as e:
    print(f"Erreur avec {filename}: {e}") pour la conversion des chunk de video en image

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.