Spaces:
Sleeping
Sleeping
File size: 11,118 Bytes
d5f8ae0 e824b96 d5f8ae0 f141fc2 4fcfcbc f141fc2 4fcfcbc f141fc2 d5f8ae0 f141fc2 d5f8ae0 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 | import re
import os
import numpy as np
import torch
import librosa
import librosa.display
import matplotlib
import matplotlib.pyplot as plt
from PIL import Image
from torchvision import transforms
import whisper
# Force non-interactive backend for server environments
matplotlib.use('Agg')
# ==========================================
# 0. Segmentation CSV Parser
# ==========================================
def parse_segmentation_csv(csv_content: bytes) -> list:
"""
Parse segmentation CSV to extract PAR speaker intervals.
CSV format: speaker,start_ms,end_ms
Returns list of (start_ms, end_ms) tuples for PAR speaker only.
"""
intervals = []
try:
lines = csv_content.decode('utf-8', errors='replace').strip().split('\n')
for i, line in enumerate(lines):
if i == 0 and 'speaker' in line.lower():
continue # Skip header
parts = line.strip().split(',')
if len(parts) >= 3 and parts[0].strip().upper() == 'PAR':
start_ms = int(parts[1].strip())
end_ms = int(parts[2].strip())
intervals.append((start_ms, end_ms))
except Exception as e:
print(f"Error parsing segmentation CSV: {e}")
return intervals
# ==========================================
# 1. Linguistic Feature Extractor
# ==========================================
class LinguisticFeatureExtractor:
def __init__(self):
self.patterns = {
'fillers': re.compile(r'&-([a-z]+)', re.IGNORECASE),
'repetition': re.compile(r'\[/+\]'),
'retracing': re.compile(r'\[//\]'),
'incomplete': re.compile(r'\+[\./]+'),
'errors': re.compile(r'\[\*.*?\]'),
'pauses': re.compile(r'\(\.+\)')
}
def clean_for_bert(self, raw_text):
text = re.sub(r'^\*PAR:\s+', '', raw_text)
text = re.sub(r'\x15\d+_\d+\x15', '', text)
text = re.sub(r'<|>', '', text)
text = re.sub(r'\[.*?\]', '', text)
text = re.sub(r'\(\.+\)', '[PAUSE]', text)
text = text.replace('_', ' ')
text = re.sub(r'\s+', ' ', text).strip()
if text.endswith('[PAUSE]'):
text = text[:-7].strip()
return text
def get_features(self, raw_text):
stats = {
'filler_count': len(self.patterns['fillers'].findall(raw_text)),
'repetition_count': len(self.patterns['repetition'].findall(raw_text)),
'retracing_count': len(self.patterns['retracing'].findall(raw_text)),
'incomplete_count': len(self.patterns['incomplete'].findall(raw_text)),
'error_count': len(self.patterns['errors'].findall(raw_text)),
'pause_count': len(self.patterns['pauses'].findall(raw_text))
}
clean_for_stats = re.sub(r'\[.*?\]', '', raw_text)
clean_for_stats = re.sub(r'&-([a-z]+)', '', clean_for_stats)
clean_for_stats = re.sub(r'[^\w\s]', '', clean_for_stats)
words = clean_for_stats.lower().split()
stats['word_count'] = len(words)
return stats
def get_feature_vector(self, raw_text):
stats = self.get_features(raw_text)
n = stats['word_count'] if stats['word_count'] > 0 else 1
# Calculate TTR (Type-Token Ratio)
clean_for_stats = re.sub(r'\[.*?\]', '', raw_text)
clean_for_stats = re.sub(r'&-([a-z]+)', '', clean_for_stats)
clean_for_stats = re.sub(r'[^\w\s]', '', clean_for_stats)
words = clean_for_stats.lower().split()
ttr = (len(set(words)) / n) if n > 0 else 0.0
return np.array([
ttr,
stats['filler_count'] / n,
stats['repetition_count'] / n,
stats['retracing_count'] / n,
stats['error_count'] / n,
stats['pause_count'] / n
], dtype=np.float32)
def extract_key_segments(self, text, max_segments=3):
"""
Extract sentences with highest linguistic marker density.
Returns list of {text, marker_count} sorted by marker count.
"""
# Split into segments using multiple delimiters:
# - Sentence endings (.?!)
# - Newlines
# - Timestamp markers (common in CHA files)
segments = re.split(r'[.?!\n]+|\x15\d+_\d+\x15', text)
segments = [s.strip() for s in segments if s.strip()]
# If no segments found, try splitting by long spaces or just use the whole text
if not segments and text.strip():
# Split by multiple spaces or use chunks of ~50 words
words = text.split()
if len(words) > 15:
# Create chunks of ~15 words each
chunk_size = 15
segments = [' '.join(words[i:i+chunk_size]) for i in range(0, len(words), chunk_size)]
else:
segments = [text.strip()]
scored = []
for sent in segments:
# Count markers in each segment
count = 0
count += len(self.patterns['fillers'].findall(sent))
count += len(self.patterns['repetition'].findall(sent))
count += len(self.patterns['retracing'].findall(sent))
count += len(self.patterns['pauses'].findall(sent))
count += len(self.patterns['errors'].findall(sent))
# Also count [PAUSE] tokens from ASR
count += sent.count('[PAUSE]')
count += sent.count('[/]')
if len(sent) > 10: # Skip very short fragments
scored.append({"text": sent, "marker_count": count})
# Sort by marker count descending
scored.sort(key=lambda x: x['marker_count'], reverse=True)
return scored[:max_segments]
# ==========================================
# 2. Audio Processor
# ==========================================
class AudioProcessor:
def __init__(self):
self.vit_transforms = transforms.Compose([
transforms.Resize((224, 224)),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])
def create_spectrogram_tensor(self, audio_path, intervals=None):
"""
Generates spectrogram image and transforms it to Tensor.
"""
try:
fig = plt.figure(figsize=(2.24, 2.24), dpi=100)
ax = fig.add_subplot(1, 1, 1)
fig.subplots_adjust(left=0, right=1, bottom=0, top=1)
if intervals:
# Load full audio then slice based on timestamps
y, sr = librosa.load(audio_path, sr=None)
clips = []
for start_ms, end_ms in intervals:
start_sample = int(start_ms * sr / 1000)
end_sample = int(end_ms * sr / 1000)
if end_sample > len(y): end_sample = len(y)
if start_sample < len(y):
clips.append(y[start_sample:end_sample])
if clips:
y = np.concatenate(clips)
else:
y = np.zeros(int(sr*30))
# Limit to 30s
if len(y) > 30 * sr:
y = y[:30 * sr]
else:
y, sr = librosa.load(audio_path, duration=30)
ms = librosa.feature.melspectrogram(y=y, sr=sr)
log_ms = librosa.power_to_db(ms, ref=np.max)
librosa.display.specshow(log_ms, sr=sr, ax=ax)
# Save to buffer instead of file
from io import BytesIO
buf = BytesIO()
fig.savefig(buf, format='png')
plt.close(fig)
buf.seek(0)
image = Image.open(buf).convert('RGB')
return self.vit_transforms(image).unsqueeze(0)
except Exception as e:
print(f"Spectrogram creation failed: {e}")
return torch.zeros((1, 3, 224, 224))
def create_spectrogram_base64(self, audio_path, intervals=None):
"""
Generates spectrogram and returns as base64 string for visualization.
"""
import base64
from io import BytesIO
try:
fig = plt.figure(figsize=(4, 3), dpi=100)
ax = fig.add_subplot(1, 1, 1)
if intervals:
y, sr = librosa.load(audio_path, sr=None)
clips = []
for start_ms, end_ms in intervals:
start_sample = int(start_ms * sr / 1000)
end_sample = int(end_ms * sr / 1000)
if end_sample > len(y): end_sample = len(y)
if start_sample < len(y):
clips.append(y[start_sample:end_sample])
if clips:
y = np.concatenate(clips)
else:
y = np.zeros(int(sr*30))
if len(y) > 30 * sr:
y = y[:30 * sr]
else:
y, sr = librosa.load(audio_path, duration=30)
ms = librosa.feature.melspectrogram(y=y, sr=sr)
log_ms = librosa.power_to_db(ms, ref=np.max)
img = librosa.display.specshow(log_ms, sr=sr, x_axis='time', y_axis='mel', ax=ax)
fig.colorbar(img, ax=ax, format='%+2.0f dB')
ax.set_title('Mel-Spectrogram')
buf = BytesIO()
fig.savefig(buf, format='png', bbox_inches='tight')
plt.close(fig)
buf.seek(0)
b64_str = base64.b64encode(buf.read()).decode('utf-8')
return f"data:image/png;base64,{b64_str}"
except Exception as e:
print(f"Spectrogram base64 creation failed: {e}")
return None
# ==========================================
# 3. ASR Helper (Whisper + CHAT Rules)
# ==========================================
def apply_chat_rules(transcription_result):
"""
Converts Whisper result into CHAT-like format AND inserts [PAUSE] tokens.
"""
formatted_text = []
segments = transcription_result.get('segments', [])
last_end = 0
for seg in segments:
gap = seg['start'] - last_end
# Insert [PAUSE] token + CHAT marker
if gap > 0.8:
formatted_text.append("[PAUSE] (..)")
elif gap > 0.3:
formatted_text.append("[PAUSE] (.)")
text = seg['text'].strip()
# Repetitions (Basic Detection)
words = text.split()
processed_words = []
for i, w in enumerate(words):
clean_w = re.sub(r'[^a-zA-Z]', '', w.lower())
if i > 0:
prev_clean = re.sub(r'[^a-zA-Z]', '', words[i-1].lower())
if clean_w == prev_clean and clean_w:
processed_words[-1] = f"{words[i-1]} [/]"
processed_words.append(w)
formatted_text.append(" ".join(processed_words))
last_end = seg['end']
return " ".join(formatted_text) |