Spaces:
Sleeping
Sleeping
File size: 3,481 Bytes
e785be9 | 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 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 |
import os
import shutil
import json
import random
from pathlib import Path
# Paths
BASE_DIR = Path("Convolve/photo/voxceleb_data")
AUDIO_SRC_DIR = BASE_DIR / "indian_celebs_audio (1)"
FACES_SRC_DIR = BASE_DIR # The _local folders are directly in BASE_DIR
# Rich Metadata Generators
RELATIONS = ["Friend", "Colleague", "Mentor", "Acquaintance", "Family", "Rival", "Partner"]
MEMORIES = [
"Met at a tech conference in Mumbai.",
"Collaborated on a creative project in 2024.",
"Shared a coffee and discussed future of AI.",
"Went to the same university.",
"Worked together on a film set.",
"Met during a charity event.",
"Regular tennis partner on weekends.",
"Discussed philosophy during a long flight.",
"Childhood neighbor from Delhi."
]
def normalize_name(folder_name):
# Removes _local, _idXXXX
name = folder_name.replace("_local", "")
if "_id" in name:
name = name.split("_id")[0]
return name
def merge_data():
print("🚀 Starting Data Merge & Enrichment...")
# Track all unique persons
persons = set()
# 1. Scan Face Folders
face_folders = [f for f in FACES_SRC_DIR.iterdir() if f.is_dir() and "_local" in f.name]
for f in face_folders:
persons.add(normalize_name(f.name))
# 2. Scan Audio Folders
if AUDIO_SRC_DIR.exists():
audio_folders = [f for f in AUDIO_SRC_DIR.iterdir() if f.is_dir()]
for f in audio_folders:
persons.add(normalize_name(f.name))
else:
print("⚠️ Audio directory not found!")
audio_folders = []
print(f"👥 Found {len(persons)} unique persons.")
# 3. Create Unified Folders
for p_name in persons:
target_dir = BASE_DIR / p_name
target_dir.mkdir(exist_ok=True)
print(f"Processing {p_name}...")
# Merge Face Data
face_src = BASE_DIR / f"{p_name}_local"
if face_src.exists():
for item in face_src.iterdir():
if item.suffix.lower() in ['.jpg', '.jpeg', '.png']:
shutil.copy2(item, target_dir / item.name)
# Remove old folder? Or keep backup? User said "proceed now", implies robust structure.
# I'll keep old for safety for now, or delete if name collision isn't issue.
# Actually, renaming might be cleaner?
# copy is safer.
# Merge Audio Data
# Audio folder name might be p_name + "_idXXXX"
# Find matching folder in AUDIO_SRC_DIR
matching_audio = [f for f in audio_folders if normalize_name(f.name) == p_name]
if matching_audio:
a_src = matching_audio[0] # Take first match
for item in a_src.iterdir():
if item.suffix.lower() == '.wav':
shutil.copy2(item, target_dir / item.name)
# Generate Rich Metadata
metadata = {
"name": p_name.replace("_", " "),
"id": p_name, # Simple ID
"relation": random.choice(RELATIONS),
"summary": random.choice(MEMORIES),
"context": {
"last_seen": "2025-12-01",
"notes": f"Generated profile for {p_name}"
}
}
# Save Metadata
with open(target_dir / "metadata.json", "w") as f:
json.dump(metadata, f, indent=4)
print("✨ Merge Complete. Data structured.")
if __name__ == "__main__":
merge_data()
|