File size: 13,240 Bytes
5806e12 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 |
import json
import os
import re
# read transcription_cuint.json from session_data/session_id
# read text_token item, repetition is start from <REPSTART> and end at <REPEND>, the
# extract repetition information and add them to current json file, information include
# content: the content between <REPSTART> and <REPEND>
# words: words index of each word in content, the index of first word in the segment is 0, index calculation ignore special token
# mark_location: the index of last words of content
# format example:
"""
"repetitions": [
{
"content": "now the now the",
"words": [
1,
2,
3,
4
],
"mark_location": 4
}
],
"""
def annotate_repetition_for_mazewhisper(session_id):
file_path = f"session_data/{session_id}/transcription_cunit.json"
if not os.path.exists(file_path):
print(f"File not found: {file_path}")
return
with open(file_path, "r", encoding="utf-8") as f:
data = json.load(f)
for segment in data["segments"]:
text_token = segment.get("text_token", "")
tokens = text_token.split()
repetitions_list = []
inside = False
rep_words = []
rep_word_indices = []
word_count = 0 # segment-level word counter
for tok in tokens:
if tok == "<REPSTART>":
inside = True
rep_words, rep_word_indices = [], []
continue
elif tok == "<REPEND>":
inside = False
if rep_words:
clean_rep_words = [w for w in rep_words
if not (w.startswith("<") and w.endswith(">"))]
repetitions_list.append(
{
"content": " ".join(clean_rep_words),
"words": rep_word_indices.copy(),
"mark_location": rep_word_indices[-1],
}
)
continue
if tok.startswith("<") and tok.endswith(">"):
if inside:
pass
else:
if inside:
rep_words.append(tok)
rep_word_indices.append(word_count)
word_count += 1
if repetitions_list:
segment["repetitions"] = repetitions_list
with open(file_path, "w", encoding="utf-8") as f:
json.dump(data, f, indent=2, ensure_ascii=False)
# read transcription_cuint.json from session_data/session_id
# read text_token item, revision is start from <REVSTART> and end at <REVEND>, the
# extract revision information and add them to current json file, information include
# content: the content between <REVSTART> and <REVEND>
# words: words index of each word in content, the index of first word in the segment is 0, index calculation ignore special token
# mark_location: the index of last words of content
# format example:
"""
"repetitions": [
{
"content": "now the now the",
"words": [
1,
2,
3,
4
],
"mark_location": 4
}
],
"""
def annotate_revision_for_mazewhisper(session_id):
file_path = f"session_data/{session_id}/transcription_cunit.json"
if not os.path.exists(file_path):
print(f"File not found: {file_path}")
return
with open(file_path, "r", encoding="utf-8") as f:
data = json.load(f)
for segment in data["segments"]:
text_token = segment.get("text_token", "")
tokens = text_token.split()
revisions_list = []
inside = False
rev_words = []
rev_word_indices = []
word_count = 0
for tok in tokens:
if tok == "<REVSTART>":
inside = True
rev_words, rev_word_indices = [], []
continue
elif tok == "<REVEND>":
inside = False
if rev_words:
clean_rev_words = [w for w in rev_words
if not (w.startswith("<") and w.endswith(">"))]
revisions_list.append(
{
"content": " ".join(clean_rev_words),
"words": rev_word_indices.copy(),
"mark_location": rev_word_indices[-1],
}
)
continue
if tok.startswith("<") and tok.endswith(">"):
continue
else:
if inside:
rev_words.append(tok)
rev_word_indices.append(word_count)
word_count += 1
if revisions_list:
segment["revisions"] = revisions_list
with open(file_path, "w", encoding="utf-8") as f:
json.dump(data, f, indent=2, ensure_ascii=False)
# read transcription_cuint.json from session_data/session_id
# read text_token item, pause is represent as <PAUSE>
# extract pause information and add them to current json file, information include
# start: the start of pause is the end timestamp of previous non-special token.
# if no previous non-special token, start time is the end time of previous segment, then set the start time of current segment as the the end time of previous segment.
# end: the end of pause is the start timestamp of next non-special token.
# if no next non-special token, then move the <PAUSE> and all possible special token after <PAUSE> to the start of next segment.
# duration: the pause duration calculated by start and end.
# format example:
"""
"pauses": [
{
"start": 364.08,
"end": 369.1,
"duration": 5.02
},
{
"start": 369.18,
"end": 369.56,
"duration": 0.38
}
],
"""
def annotate_pause_for_mazewhisper(session_id):
file_path = f"session_data/{session_id}/transcription_cunit.json"
if not os.path.exists(file_path):
print(f"File not found: {file_path}")
return
with open(file_path, 'r', encoding='utf-8') as f:
data = json.load(f)
for i, segment in enumerate(data['segments']):
text_token = segment.get('text_token', '')
words = segment.get('words', [])
pauses_list = []
if '<PAUSE>' in text_token:
tokens = text_token.split()
for j, token in enumerate(tokens):
if token == '<PAUSE>':
# Find start time: end timestamp of previous non-special token
start_time = None
# Look backwards in current segment for previous word
word_idx = 0
for k in range(j):
if not (tokens[k].startswith('<') and tokens[k].endswith('>')):
word_idx += 1
if word_idx > 0 and word_idx <= len(words):
start_time = words[word_idx - 1].get('end')
elif i > 0: # Use end time of previous segment
start_time = data['segments'][i - 1].get('end')
# Update current segment start time
segment['start'] = start_time
# Find end time: start timestamp of next non-special token
end_time = None
# Look forwards in current segment for next word
next_word_idx = 0
for k in range(j + 1, len(tokens)):
if not (tokens[k].startswith('<') and tokens[k].endswith('>')):
next_word_idx = word_idx + 1
break
if tokens[k].startswith('<') and tokens[k].endswith('>'):
continue
if next_word_idx > 0 and next_word_idx <= len(words):
end_time = words[next_word_idx - 1].get('start')
elif i < len(data['segments']) - 1: # Move pause to next segment
# This is complex - for now, use next segment start
end_time = data['segments'][i + 1].get('start') if i + 1 < len(data['segments']) else segment.get('end')
# If we have both start and end times, calculate duration
if start_time is not None and end_time is not None:
duration = round(end_time - start_time, 2)
pause_info = {
"start": start_time,
"end": end_time,
"duration": duration
}
pauses_list.append(pause_info)
# Add pauses to this segment if any found
if pauses_list:
segment['pauses'] = pauses_list
# Save the updated data back to file
with open(file_path, 'w', encoding='utf-8') as f:
json.dump(data, f, indent=2, ensure_ascii=False)
return
# read transcription_cuint.json from session_data/session_id
# read text_token item, filler words is represent as <FILLER>
# extract filler word information and add them to current json file, information include
# start: set to empty for now
# end: set to empty for now
# duration: set to empty for now
# format example:
"""
"fillerwords": [
{
"start": ,
"end": ,
"content": "",
"duration":
}
],
"""
# then, insert "um" as the previous token of <FILLER> in text_token item.
# then, insert "um" to text item and word list (timestamp set to empty for now), the location is corresponding to the <FILLER> location in text_token
# (this is a prelimary implementation, next steps should use maze whisper that transcript filler words content and align them directly)
def annotate_fillerword_for_mazewhisper(session_id):
file_path = f"session_data/{session_id}/transcription_cunit.json"
if not os.path.exists(file_path):
print(f"File not found: {file_path}")
return
with open(file_path, 'r', encoding='utf-8') as f:
data = json.load(f)
for segment in data['segments']:
text_token = segment.get('text_token', '')
words = segment.get('words', [])
text = segment.get('text', '')
fillerwords_list = []
if '<FILLER>' in text_token:
# Create filler word entries (preliminary implementation)
filler_count = text_token.count('<FILLER>')
for _ in range(filler_count):
filler_info = {
"start": None,
"end": None,
"content": "",
"duration": None
}
fillerwords_list.append(filler_info)
# Insert "um" before each <FILLER> in text_token
updated_text_token = text_token.replace('<FILLER>', 'um <FILLER>')
segment['text_token'] = updated_text_token
# Insert "um" in text and words list at corresponding locations
tokens = text_token.split()
text_words = text.split()
# Find positions of <FILLER> and insert "um"
filler_positions = []
word_count = 0
for i, token in enumerate(tokens):
if token == '<FILLER>':
filler_positions.append(word_count)
elif not (token.startswith('<') and token.endswith('>')):
word_count += 1
# Insert "um" in reverse order to maintain indices
for pos in reversed(filler_positions):
text_words.insert(pos, 'um')
# Insert in words list with empty timestamps
um_word = {
"word": "um",
"start": None,
"end": None
}
if pos <= len(words):
words.insert(pos, um_word)
# Update text and words in segment
segment['text'] = ' '.join(text_words)
segment['words'] = words
# Add fillerwords to this segment
segment['fillerwords'] = fillerwords_list
# Save the updated data back to file
with open(file_path, 'w', encoding='utf-8') as f:
json.dump(data, f, indent=2, ensure_ascii=False)
return
def annotate_maze_for_mazewhisper(session_id):
annotate_fillerword_for_mazewhisper(session_id)
annotate_repetition_for_mazewhisper(session_id)
annotate_revision_for_mazewhisper(session_id)
annotate_pause_for_mazewhisper(session_id)
print("Maze annotation completed!")
|