initial commit

.gitignore

@@ -0,0 +1,3 @@
+*.png
+*.flac
+*.wav

create_chunks.py

@@ -0,0 +1,52 @@
+#!/usr/bin/env python3
+import os
+import subprocess
+import datetime
+import pytz
+
+OUTDIR = "chunks_unprocessed"
+DURATION = 10800  # 3 Stunden
+SOURCE = "pulse"  # funktioniert mit PulseAudio
+
+os.makedirs(OUTDIR, exist_ok=True)
+
+def get_next_section_start(now):
+    hour = now.hour
+    next_boundary = (hour // 3 + 1) * 3
+    if next_boundary >= 24:
+        next_time = now.replace(day=now.day, hour=0, minute=0, second=0, microsecond=0) + datetime.timedelta(days=1)
+    else:
+        next_time = now.replace(hour=next_boundary, minute=0, second=0, microsecond=0)
+    return next_time
+
+def record_section():
+    tz = pytz.timezone("Europe/Berlin")
+    now = datetime.datetime.now(tz)
+    timestamp = now.strftime("%Y%m%d-%H%M")
+    filename = os.path.join(OUTDIR, f"{timestamp}.flac")
+    print(f"🎙️ Starte Aufnahme: {filename}")
+
+    next_start = get_next_section_start(now)
+    duration_seconds = int((next_start - now).total_seconds())
+
+    cmd = [
+        "ffmpeg",
+        "-f", "pulse",
+        "-i", "alsa_input.usb-HANMUS_USB_AUDIO_24BIT_2I2O_1612310-00.analog-stereo",
+        "-ac", "1",
+        "-ar", "96000",
+        "-sample_fmt", "s32",
+        "-t", str(duration_seconds),
+        "-c:a", "flac",
+        "-compression_level", "12",
+        filename
+    ]
+    subprocess.run(cmd)
+    print(f"✅ Aufnahme abgeschlossen: {filename}")
+
+def main():
+    while True:
+        record_section()
+
+if __name__ == "__main__":
+    main()

process_chunks.py

@@ -0,0 +1,107 @@
+#!/usr/bin/env python3
+"""
+Erkennt 211 Hz + 422 Hz (Oberton) in WAV-Dateien.
+Speichert WAV + PNG nur bei Erkennung.
+Blockiert Folgetreffer für definierte Zeit (SKIP_SECONDS).
+"""
+
+import numpy as np
+import soundfile as sf
+from scipy.fft import fft, fftfreq
+import matplotlib.pyplot as plt
+import os
+
+# === Konfiguration ===
+FILENAME = "1b.flac"
+TARGET_FREQ = 211
+OCTAVE_FREQ = TARGET_FREQ * 2
+TOLERANCE = 1
+THRESHOLD_BASE = 0.3
+THRESHOLD_OCT = THRESHOLD_BASE / 10
+CHUNK_SECONDS = 2
+CLIP_PADDING_BEFORE = 2
+CLIP_PADDING_AFTER = 8
+SKIP_SECONDS = 10
+OUTDIR = "events"
+
+os.makedirs(OUTDIR, exist_ok=True)
+
+# === WAV/Audio-Datei laden ===
+data, rate = sf.read(FILENAME, dtype='float32')
+if data.ndim > 1:
+    data = data.mean(axis=1)
+
+samples_per_chunk = int(rate * CHUNK_SECONDS)
+total_chunks = len(data) // samples_per_chunk
+
+detections = []
+next_allowed_time = 0  # für Skip-Logik
+
+# === Analyse-Loop ===
+for i in range(total_chunks):
+    timestamp = i * CHUNK_SECONDS
+    if timestamp < next_allowed_time:
+        continue
+
+    segment = data[i * samples_per_chunk : (i + 1) * samples_per_chunk]
+    if len(segment) == 0:
+        continue
+
+    freqs = fftfreq(len(segment), d=1/rate)
+    fft_vals = np.abs(fft(segment))
+
+    pos_mask = freqs > 0
+    freqs = freqs[pos_mask]
+    fft_vals = fft_vals[pos_mask]
+
+    peak_freq = freqs[np.argmax(fft_vals)]
+    peak_mag = np.max(fft_vals)
+
+    # Energien normiert
+    mask_base = (freqs >= TARGET_FREQ - TOLERANCE) & (freqs <= TARGET_FREQ + TOLERANCE)
+    energy_base = np.mean(fft_vals[mask_base]) / peak_mag
+
+    mask_oct = (freqs >= OCTAVE_FREQ - TOLERANCE) & (freqs <= OCTAVE_FREQ + TOLERANCE)
+    energy_oct = np.mean(fft_vals[mask_oct]) / peak_mag
+
+    is_peak_near_target = TARGET_FREQ - TOLERANCE <= peak_freq <= TARGET_FREQ + TOLERANCE
+    detected = is_peak_near_target and energy_base > THRESHOLD_BASE and energy_oct > THRESHOLD_OCT
+
+    if detected:
+        detections.append((timestamp, round(energy_base, 4), round(energy_oct, 4), round(peak_freq, 2)))
+        next_allowed_time = timestamp + SKIP_SECONDS
+
+        # Ausschnitt extrahieren
+        start = max(0, int((timestamp - CLIP_PADDING_BEFORE) * rate))
+        end = min(len(data), int((timestamp + CLIP_PADDING_AFTER) * rate))
+        clip = (data[start:end] * 32767).astype(np.int16)
+
+        base_filename = os.path.join(OUTDIR, f"event_{int(timestamp):04}s")
+        wav_name = f"{base_filename}.wav"
+        png_name = f"{base_filename}.png"
+
+        # WAV speichern
+        sf.write(wav_name, clip, rate, subtype="PCM_24")
+        print(f"🟢 WAV gespeichert: {wav_name} (211Hz: {energy_base:.4f}, 422Hz: {energy_oct:.4f}, Peak: {peak_freq:.1f} Hz)")
+
+        # PNG Spektrogramm
+        plt.figure(figsize=(10, 4))
+        # Verstärke das Signal künstlich, um schwache Ereignisse im dB-Spektrum sichtbarer zu machen
+        plt.specgram((clip / 32767.0), NFFT=32768, Fs=rate, noverlap=512, cmap="plasma", vmin=-80, vmax=-35)
+        plt.title(f"Ereignis @ {timestamp:.2f}s")
+        plt.xlabel("Zeit (s)")
+        plt.ylabel("Frequenz (Hz)")
+        plt.ylim(0, 1000)
+        plt.colorbar(label="Intensität (dB)")
+        plt.tight_layout()
+        plt.savefig(png_name)
+        plt.close()
+        print(f"📷 PNG gespeichert: {png_name}")
+
+# === Zusammenfassung ===
+print("\n🎯 Erkennungen:")
+for ts, eb, eo, pf in detections:
+    print(f"- {ts:.2f}s | 211Hz: {eb} | 422Hz: {eo} | Peak: {pf:.1f} Hz")
+
+if not detections:
+    print("→ Keine gültigen Ereignisse erkannt.")