monkeycall/dat2wav.py at main · MattechLab/monkeycall · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
import os
import csv
import numpy as np
import matplotlib.pyplot as plt
from scipy.io.wavfile import write

# === Configuration ===
input_base = "/run/user/1000/gvfs/smb-share:server=filer01.hevs.ch,share=fs_projets/MatTechLab/monkey_call/00_rawdata_collars_2023"
output_base = "/run/user/1000/gvfs/smb-share:server=filer01.hevs.ch,share=fs_projets/MatTechLab/monkey_call/output"
sample_rate = 16000       # Hz
dtype = np.int16          # Signed 16-bit PCM
num_channels = 2          # 1 = mono, 2 = stereo

# === Prepare log file ===
log_path = os.path.join(output_base, "processing_log.csv")
os.makedirs(output_base, exist_ok=True)
log_entries = []

# === Recursively find .DAT files ===
for root, dirs, files in os.walk(input_base):
    for filename in files:
        if filename.lower().endswith(".dat"):
            dat_path = os.path.join(root, filename)

            # Skip empty files
            if os.path.getsize(dat_path) == 0:
                print(f"Skipping empty file: {dat_path}")
                log_entries.append([dat_path, "", "", "Empty file"])
                continue

            print(f"Processing: {dat_path}")

            # === Load the raw audio ===
            with open(dat_path, "rb") as f:
                audio_data = np.frombuffer(f.read(), dtype=dtype)

            # === Reshape if stereo ===
            if num_channels > 1:
                try:
                    audio_data = audio_data.reshape(-1, num_channels)
                except ValueError:
                    print(f"Warning: Cannot reshape file (size mismatch), skipping: {dat_path}")
                    log_entries.append([dat_path, "", "", "Reshape error"])
                    continue

            # === Determine output folder ===
            relative_path = os.path.relpath(root, input_base)
            output_folder = os.path.join(output_base, relative_path)
            os.makedirs(output_folder, exist_ok=True)

            base_name = os.path.splitext(filename)[0]
            wav_path = os.path.join(output_folder, base_name + ".wav")
            png_path = os.path.join(output_folder, base_name + ".png")

            # === Save WAV ===
            write(wav_path, sample_rate, audio_data)
            print(f"Saved WAV to: {wav_path}")

            # === Save waveform plot ===
            plt.figure(figsize=(10, 4))
            if audio_data.ndim == 1:
                plt.plot(audio_data[:1000])
                plt.title(f"{base_name} (Mono)")
            else:
                plt.plot(audio_data[:1000, 0], label="Left")
                plt.plot(audio_data[:1000, 1], label="Right")
                plt.title(f"{base_name} (Stereo)")
                plt.legend()

            plt.xlabel("Sample")
            plt.ylabel("Amplitude")
            plt.tight_layout()
            plt.savefig(png_path)
            plt.close()
            print(f"Saved waveform to: {png_path}")

            # === Log success ===
            log_entries.append([dat_path, wav_path, png_path, "OK"])

# === Write log file ===
with open(log_path, mode='w', newline='') as csvfile:
    writer = csv.writer(csvfile)
    writer.writerow(["DAT file", "WAV output", "PNG output", "Status"])
    writer.writerows(log_entries)

print(f"\n📋 Log written to: {log_path}")
print("🎉 Processing complete!")