wip

process

@@ -3,7 +3,7 @@ import os
 import datetime
 import numpy as np
 import matplotlib.pyplot as plt
-import soundfile
+import soundfile as sf
 from scipy.fft import rfft, rfftfreq
 import shutil
 import traceback
@@ -36,20 +36,20 @@ def process_recording(filename):
 
     # get data and metadata from recording
     path = os.path.join(RECORDINGS_DIR, filename)
-    sound, samplerate = soundfile.read(path)
+    soundfile = sf.SoundFile(path)
+    samplerate = soundfile.samplerate
     samples_per_block = int(BLOCK_SECONDS * samplerate)
     overlapping_samples = int(samples_per_block * BLOCK_OVERLAP_FACTOR)
 
-    # chache data about current event
+    sample_num = 0
     current_event = None
 
-    # read blocks of audio data with overlap from sound variable
+    while sample_num < len(soundfile):
-    sample_num = 0
+        soundfile.seek(sample_num)
-    while sample_num < len(sound):
+        block = soundfile.read(frames=samples_per_block, dtype='float32', always_2d=False)
-        # get block of audio data
+
-        block_start = sample_num
+        if len(block) == 0:
-        block_end = min(sample_num + samples_per_block, len(sound))
+            break
-        block = sound[block_start:block_end]
 
         # calculate FFT
         labels = rfftfreq(len(block), d=1/samplerate)
@@ -102,7 +102,8 @@ def process_recording(filename):
                 current_event['duration'] = duration
                 print(f'🔊 {current_event['start_at'].strftime('%Y-%m-%d %H:%M:%S')} ({duration:.1f}s): {current_event['start_freq']:.1f}Hz->{current_event['end_freq']:.1f}Hz @{current_event['max_amplitude']:.0f}rDB')
 
-                write_event(current_event=current_event, sound=sound, samplerate=samplerate)
+                # read full audio clip again for writing
+                write_event(current_event=current_event, soundfile=soundfile, samplerate=samplerate)
 
                 current_event = None
                 sample_num += DETECTION_DISTANCE_BLOCKS * samples_per_block
@@ -111,27 +112,31 @@ def process_recording(filename):
 
 
 # write a spectrogram using the sound from start to end of the event
-def write_event(current_event, sound, samplerate):
+def write_event(current_event, soundfile, samplerate):
+    # date and filename
+    event_date = current_event['start_at'] - datetime.timedelta(seconds=PLOT_PADDING_START_SECONDS)
+    filename_prefix = event_date.strftime('%Y-%m-%d_%H-%M-%S.%f%z')
+
+    # event clip
     event_start_sample = current_event['start_sample'] - samplerate * PLOT_PADDING_START_SECONDS
     event_end_sample = current_event['end_sample'] + samplerate * PLOT_PADDING_END_SECONDS
-    event_clip = sound[event_start_sample:event_end_sample]
+    total_samples = event_end_sample - event_start_sample
-    event = current_event['start_at'] - datetime.timedelta(seconds=PLOT_PADDING_START_SECONDS)
+    soundfile.seek(event_start_sample)
-    filename_prefix = current_event['start_at'].strftime('%Y-%m-%d_%H-%M-%S.%f%z')
+    event_clip = soundfile.read(frames=total_samples, dtype='float32', always_2d=False)
 
     # write flac
     flac_path = os.path.join(DETECTIONS_DIR, f"{filename_prefix}.flac")
-    soundfile.write(flac_path, event_clip, samplerate, format='FLAC')
+    sf.write(flac_path, event_clip, samplerate, format='FLAC')
 
     # write spectrogram
     plt.figure(figsize=(8, 6))
     plt.specgram(event_clip, Fs=samplerate, NFFT=samplerate, noverlap=samplerate//2, cmap='inferno', vmin=-100, vmax=-10)
-    plt.title(f"Bootshorn @{event.strftime('%Y-%m-%d %H:%M:%S%z')}")
+    plt.title(f"Bootshorn @{event_date.strftime('%Y-%m-%d %H:%M:%S%z')}")
     plt.xlabel(f"Time {current_event['duration']:.1f}s")
     plt.ylabel(f"Frequency {current_event['start_freq']:.1f}Hz -> {current_event['end_freq']:.1f}Hz")
     plt.colorbar(label="Intensity (rDB)")
     plt.ylim(50, 1000)
-    spectrogram_path = os.path.join(DETECTIONS_DIR, f"{filename_prefix}.png")
+    plt.savefig(os.path.join(DETECTIONS_DIR, f"{filename_prefix}.png"))
-    plt.savefig(spectrogram_path)
     plt.close()