Upload folder using huggingface_hub
Browse files- LEVI_whisper_benchmark.py +47 -0
- __init__.py +0 -0
- __pycache__/__init__.cpython-310.pyc +0 -0
- __pycache__/benchmark_utils.cpython-310.pyc +0 -0
- __pycache__/converters.cpython-310.pyc +0 -0
- __pycache__/renamers.cpython-310.pyc +0 -0
- __pycache__/trimmers.cpython-310.pyc +0 -0
- benchmark_utils.py +353 -0
- converters.py +206 -0
- renamers.py +77 -0
- trimmers.py +139 -0
LEVI_whisper_benchmark.py
ADDED
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
#%% imports
|
2 |
+
import os
|
3 |
+
from benchmark_utils import ASRmanifest, wer_from_csv
|
4 |
+
|
5 |
+
|
6 |
+
#%% setup paths
|
7 |
+
corpora_root = '/shared/corpora/forSAGA/' # root path where audio files are, inserted in palce of $DATAROOT in manifest
|
8 |
+
manif_root = '/shared/corpora/forSAGA/data_manifests/' # path to dir containing data manifest csvs
|
9 |
+
output_dir = './ASR_output/' # where to save ASR output
|
10 |
+
manifest='LEVI_LoFi_v2_TEST_norm_wer_isat' # name of test manifest
|
11 |
+
model_name= 'LEVI_whisper_medium.en' # name of save directory of model you want to evaluate
|
12 |
+
hf_org = 'levicu'
|
13 |
+
model_path = f'{hf_org}/{model_name}'
|
14 |
+
|
15 |
+
#%% setup paths for Rosy TESTING:
|
16 |
+
corpora_root = '/shared/corpora/' # root path where audio files are, inserted in palce of $DATAROOT in manifest
|
17 |
+
manif_root = '/shared/corpora/data_manifests/ASR/' # path to dir containing data manifest csvs
|
18 |
+
output_dir = '/home/rosy/whisat-output/' # where to save ASR output
|
19 |
+
manifest= 'LEVI_LoFi_v2_TEST_punc+cased' # name of test manifest
|
20 |
+
model_name= 'LEVI_LoFi_v2_MediumEN_Lora_Int8' # name of save directory of model you want to evaluate
|
21 |
+
model_path='/shared/models/LEVI_LoFi_v2_MediumEN_Lora_Int8/final/'
|
22 |
+
model_path='openai/whisper_medium.en'
|
23 |
+
#%%
|
24 |
+
# generate paths
|
25 |
+
manifest_csv=os.path.join(manif_root, f'{manifest}.csv')
|
26 |
+
out_csv=os.path.join(output_dir,f'{model_name}_on_{manifest}.csv')
|
27 |
+
|
28 |
+
#%% Inference
|
29 |
+
ASRmanifest(
|
30 |
+
manifest_csv=manifest_csv,
|
31 |
+
out_csv=out_csv,
|
32 |
+
corpora_root=corpora_root,
|
33 |
+
model_path=model_path,
|
34 |
+
)
|
35 |
+
|
36 |
+
#%% Evaluation
|
37 |
+
print(f'reading results from {out_csv}')
|
38 |
+
print(f'{model_name} on {manifest}')
|
39 |
+
wer_meas=wer_from_csv(
|
40 |
+
out_csv,
|
41 |
+
refcol='transcript',
|
42 |
+
hypcol='asr',
|
43 |
+
printout=True,
|
44 |
+
text_norm_method='levi'
|
45 |
+
)
|
46 |
+
|
47 |
+
|
__init__.py
ADDED
File without changes
|
__pycache__/__init__.cpython-310.pyc
ADDED
Binary file (128 Bytes). View file
|
|
__pycache__/benchmark_utils.cpython-310.pyc
ADDED
Binary file (10.2 kB). View file
|
|
__pycache__/converters.cpython-310.pyc
ADDED
Binary file (5.32 kB). View file
|
|
__pycache__/renamers.cpython-310.pyc
ADDED
Binary file (1.61 kB). View file
|
|
__pycache__/trimmers.cpython-310.pyc
ADDED
Binary file (2.77 kB). View file
|
|
benchmark_utils.py
ADDED
@@ -0,0 +1,353 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
#%% imports
|
2 |
+
import pandas as pd
|
3 |
+
import time
|
4 |
+
from tqdm import tqdm
|
5 |
+
import torch
|
6 |
+
from torch.cuda.amp import autocast
|
7 |
+
import transformers
|
8 |
+
from transformers import WhisperFeatureExtractor, WhisperTokenizer, WhisperProcessor, WhisperForConditionalGeneration, GenerationConfig
|
9 |
+
from transformers import pipeline, AutomaticSpeechRecognitionPipeline
|
10 |
+
from peft import PeftModel, PeftConfig
|
11 |
+
import warnings
|
12 |
+
import jiwer
|
13 |
+
from jiwer.process import WordOutput
|
14 |
+
import pandas as pd
|
15 |
+
import numpy as np
|
16 |
+
from pathlib import Path
|
17 |
+
import os
|
18 |
+
import math
|
19 |
+
from decimal import InvalidOperation
|
20 |
+
import contractions
|
21 |
+
from whisper.normalizers.english import EnglishTextNormalizer
|
22 |
+
from num2words import num2words
|
23 |
+
import csv
|
24 |
+
import re
|
25 |
+
import string
|
26 |
+
|
27 |
+
#%% define functions
|
28 |
+
def ASRmanifest(
|
29 |
+
manifest_csv: str,
|
30 |
+
out_csv: str,
|
31 |
+
corpora_root: str,
|
32 |
+
model_path:str,
|
33 |
+
):
|
34 |
+
"""Run Whisper ASR on a dataset specified in a manifest
|
35 |
+
Args:
|
36 |
+
manifest_csv (str): path to manifest csv listing files to transcribe
|
37 |
+
out_csv (str):path to write output csv
|
38 |
+
corpora_root (str): root path where audio files are, inserted in place of $DATAROOT in manifest
|
39 |
+
model_path (str): path to model directory / huggingface model name
|
40 |
+
"""
|
41 |
+
|
42 |
+
df = pd.read_csv(manifest_csv,keep_default_na=False)
|
43 |
+
fieldnames = list(df.columns) + ['asr']
|
44 |
+
|
45 |
+
asr_pipeline=prepare_pipeline(
|
46 |
+
model_path=model_path,
|
47 |
+
generate_opts={'max_new_tokens':448,
|
48 |
+
'num_beams':1,#greedy
|
49 |
+
'repetition_penalty':1,
|
50 |
+
'do_sample':False
|
51 |
+
}
|
52 |
+
)
|
53 |
+
|
54 |
+
message = "This may take a while on CPU." if asr_pipeline.device.type=="cpu" else "Using GPU"
|
55 |
+
print(f'Running ASR for {len(df)} files. {message} ...')
|
56 |
+
compute_time=0
|
57 |
+
total_audio_dur=0
|
58 |
+
# get the start time
|
59 |
+
st = time.time()
|
60 |
+
|
61 |
+
with open(out_csv, 'w', newline='') as csvfile:
|
62 |
+
writer = csv.DictWriter(csvfile, fieldnames=fieldnames,delimiter=',')
|
63 |
+
writer.writeheader()
|
64 |
+
for i,row in tqdm(df.iterrows(), total=df.shape[0]):
|
65 |
+
audiofile=row['wav'].replace('$DATAROOT',corpora_root)
|
66 |
+
with torch.no_grad():
|
67 |
+
with autocast():
|
68 |
+
try:
|
69 |
+
result = asr_pipeline(audiofile)
|
70 |
+
asrtext = result['text']
|
71 |
+
except (FileNotFoundError, ValueError) as e:
|
72 |
+
print(f'SKIPPED: {audiofile}')
|
73 |
+
continue
|
74 |
+
row['asr']=asrtext
|
75 |
+
writer.writerow( row.to_dict())
|
76 |
+
et = time.time()
|
77 |
+
compute_time = (et-st)
|
78 |
+
print(f'...transcription complete in {compute_time:.1f} sec')
|
79 |
+
|
80 |
+
def load_model(
|
81 |
+
model_path:str,
|
82 |
+
language='english',
|
83 |
+
use_int8 = False,
|
84 |
+
device_map='auto'):
|
85 |
+
|
86 |
+
warnings.filterwarnings("ignore")
|
87 |
+
transformers.utils.logging.set_verbosity_error()
|
88 |
+
|
89 |
+
try:
|
90 |
+
model = WhisperForConditionalGeneration.from_pretrained(
|
91 |
+
model_path,
|
92 |
+
load_in_8bit=use_int8,
|
93 |
+
device_map=device_map,
|
94 |
+
use_cache=False,
|
95 |
+
)
|
96 |
+
try:
|
97 |
+
processor=WhisperProcessor.from_pretrained(model_path, language=language, task="transcribe")
|
98 |
+
except OSError:
|
99 |
+
print('missing tokenizer and preprocessor config files in save dir, checking directory above...')
|
100 |
+
processor=WhisperProcessor.from_pretrained(os.path.join(model_path,'..'), language=language, task="transcribe")
|
101 |
+
|
102 |
+
except OSError as e:
|
103 |
+
print(f'{e}: possibly missing model or config file in model path. Will check for adapter...')
|
104 |
+
# check if PEFT
|
105 |
+
if os.path.isdir(os.path.join(model_path , "adapter_model")):
|
106 |
+
print('found adapter...loading PEFT model')
|
107 |
+
# checkpoint dir needs adapter model subdir with adapter_model.bin and adapter_confg.json
|
108 |
+
peft_config = PeftConfig.from_pretrained(os.path.join(model_path , "adapter_model"))
|
109 |
+
print(f'...loading and merging LORA weights to base model {peft_config.base_model_name_or_path}')
|
110 |
+
model = WhisperForConditionalGeneration.from_pretrained(peft_config.base_model_name_or_path,
|
111 |
+
load_in_8bit=use_int8,
|
112 |
+
device_map=device_map,
|
113 |
+
use_cache=False,
|
114 |
+
)
|
115 |
+
model = PeftModel.from_pretrained(model, os.path.join(model_path,"adapter_model"))
|
116 |
+
model = model.merge_and_unload()
|
117 |
+
processor = WhisperProcessor.from_pretrained(peft_config.base_model_name_or_path, language=language, task="transcribe")
|
118 |
+
else:
|
119 |
+
raise e
|
120 |
+
model.eval()
|
121 |
+
return(model, processor)
|
122 |
+
|
123 |
+
def prepare_pipeline(model_path, generate_opts):
|
124 |
+
"""Prepare a pipeline for ASR inference
|
125 |
+
Args:
|
126 |
+
model_path (str): path to model directory / huggingface model name
|
127 |
+
generate_opts (dict): options to pass to pipeline
|
128 |
+
Returns:
|
129 |
+
pipeline: ASR pipeline
|
130 |
+
"""
|
131 |
+
model, processor = load_model(
|
132 |
+
model_path=model_path)
|
133 |
+
|
134 |
+
asr_pipeline = pipeline(
|
135 |
+
"automatic-speech-recognition",
|
136 |
+
model=model,
|
137 |
+
tokenizer=processor.tokenizer,
|
138 |
+
feature_extractor=processor.feature_extractor,
|
139 |
+
generate_kwargs=generate_opts,
|
140 |
+
)
|
141 |
+
return asr_pipeline
|
142 |
+
|
143 |
+
#%% WER evaluation functions
|
144 |
+
def get_normalizer(text_norm_method='isat'):
|
145 |
+
if text_norm_method=='whisper':
|
146 |
+
normalizer=whisper_norm_text_for_wer
|
147 |
+
elif text_norm_method=='whisper_keep_tags':
|
148 |
+
normalizer=EnglishTextNormalizer()
|
149 |
+
elif text_norm_method=='isat':
|
150 |
+
normalizer = norm_text_for_wer
|
151 |
+
elif text_norm_method=='levi':
|
152 |
+
normalizer = levi_norm_text_for_wer
|
153 |
+
else:
|
154 |
+
raise NotImplementedError(f'unrecognized normalizer method: {text_norm_method}')
|
155 |
+
return normalizer
|
156 |
+
|
157 |
+
def strip_punct(instr, keep_math=False):
|
158 |
+
newstr = ''
|
159 |
+
for word in instr.split():
|
160 |
+
if keep_math:
|
161 |
+
word=word.strip('!"#$&\',.:;<=>?@[\\]^_`{|}~')
|
162 |
+
else:
|
163 |
+
# delete punct from start and end of word
|
164 |
+
word = word.strip(string.punctuation)
|
165 |
+
# delete commas inside numbers
|
166 |
+
m = re.match(r'(\d*),(\d)', word)
|
167 |
+
if m != None:
|
168 |
+
word = word.replace(',', '')
|
169 |
+
# commas inside words become space
|
170 |
+
word = re.sub(",", " ", word)
|
171 |
+
# hyphens inside words become space
|
172 |
+
if keep_math:
|
173 |
+
pass
|
174 |
+
else:
|
175 |
+
word = re.sub("-", " ", word)
|
176 |
+
word = word.strip()
|
177 |
+
newstr += ' ' + word
|
178 |
+
newstr = newstr.strip()
|
179 |
+
return newstr
|
180 |
+
|
181 |
+
def remove_in_brackets(text):
|
182 |
+
# removes any clause in brackets or parens, and the brackets themselves
|
183 |
+
return re.sub("[\(\[\<].*?[\)\]\>]+", " ", text)
|
184 |
+
|
185 |
+
def caught_num2words(text):
|
186 |
+
# first do currency replacements #TODO: plurals vs singular
|
187 |
+
if '$' in text:
|
188 |
+
text = re.sub('\$([0-9]+)', '\g<1> dollars', text)
|
189 |
+
if '€' in text:
|
190 |
+
text = re.sub('\$([0-9]+)', '\g<1> euro', text)
|
191 |
+
if '£' in text:
|
192 |
+
text = re.sub('\$([0-9]+)', '\g<1> pounds', text)
|
193 |
+
if '%' in text:
|
194 |
+
text = re.sub('([0-9]+)\%', '\g<1> percent', text)
|
195 |
+
|
196 |
+
# strip punctuation
|
197 |
+
text=strip_punct(text, keep_math=True)
|
198 |
+
text=text.strip('*=/')
|
199 |
+
# catch strings that might be converted to infinity or NaN and return as is...
|
200 |
+
naughty_words = ['INF','Inf','inf','NAN','NaN', 'nan', 'NONE','None','none','Infinity','infinity']
|
201 |
+
if text in naughty_words:
|
202 |
+
return text
|
203 |
+
try:
|
204 |
+
if len(text.split()) > 1:
|
205 |
+
return ' '.join([caught_num2words(word) for word in text.split()])
|
206 |
+
else:
|
207 |
+
return num2words(text)
|
208 |
+
except (InvalidOperation, ValueError) as error:
|
209 |
+
return text
|
210 |
+
|
211 |
+
def spell_math(text):
|
212 |
+
# spell out mathematical expressions
|
213 |
+
# numerals preceded by hyphen become negative
|
214 |
+
text = re.sub('\-(\d+)', 'minus \g<1>', text)
|
215 |
+
text = re.sub('(\d+\s?)\-(\s?\d?)', '\g<1> minus \g<2>', text)
|
216 |
+
text = re.sub('(\w+\s+)\-(\s?\w+)', '\g<1> minus \g<2>', text) # need to be more careful with - as this could be a hyphenated word not minus
|
217 |
+
text = re.sub('(\w+\s?)\+(\s?\w+)', '\g<1> plus \g<2>', text)
|
218 |
+
text = re.sub('(\w+\s?)\*(\s?\w+)', '\g<1> times \g<2>', text)
|
219 |
+
text = re.sub('(\d+\s?)x(\s?\d)', '\g<1> times \g<2>', text) # need to be more careful with x as this could be a variable not times
|
220 |
+
text = re.sub('(\w+\s?)\/(\s?\w+)', '\g<1> divided by \g<2>', text)
|
221 |
+
text = re.sub('(\w+\s?)\=(\s?\w+)', '\g<1> equals \g<2>', text)
|
222 |
+
return text
|
223 |
+
|
224 |
+
def expand_contractions(str):
|
225 |
+
expanded_words = []
|
226 |
+
for wrd in str.split():
|
227 |
+
expanded_words.append(contractions.fix(wrd))
|
228 |
+
str = ' '.join(expanded_words)
|
229 |
+
return str
|
230 |
+
|
231 |
+
def norm_text_for_wer(text):
|
232 |
+
# function to format text or lists of text (e.g. asr, transcript) for wer computation.
|
233 |
+
# Converts from list to a single string and apply some text normalization operations
|
234 |
+
# note that the clean_REV_transcript function should be applied first to remove REV-specific keywords
|
235 |
+
# and extract text from docx format tables
|
236 |
+
|
237 |
+
if isinstance(text,list):
|
238 |
+
text = ' '.join(text)
|
239 |
+
text=str(text)
|
240 |
+
text = text.replace('\n',' ') # replace newline with space
|
241 |
+
text = remove_in_brackets(text) # removes non-spoken annotations such as [inaudible]
|
242 |
+
text = re.sub('%\w+','', text) # remove %HESITATION etc
|
243 |
+
text = ' '.join([caught_num2words(str) for str in text.split(' ')]) # spell out numbers
|
244 |
+
text = expand_contractions(text)
|
245 |
+
text = strip_punct(text)
|
246 |
+
text = text.lower()
|
247 |
+
text = re.sub('\s+',' ',text) # replace multiple space with single
|
248 |
+
return text
|
249 |
+
|
250 |
+
def levi_norm_text_for_wer(text):
|
251 |
+
# function to format text or lists of text (e.g. asr, transcript) for wer computation.
|
252 |
+
# specialized for math language
|
253 |
+
|
254 |
+
if isinstance(text,list):
|
255 |
+
text = ' '.join(text)
|
256 |
+
text=str(text)
|
257 |
+
text = text.replace('\n',' ') # replace newline with space
|
258 |
+
text = remove_in_brackets(text) # removes non-spoken annotations such as [inaudible]
|
259 |
+
text = re.sub('%\w+','', text) # remove %HESITATION etc
|
260 |
+
text = spell_math(text)
|
261 |
+
text = ' '.join([caught_num2words(str) for str in text.split(' ')]) # spell out numbers
|
262 |
+
text = expand_contractions(text)
|
263 |
+
text = strip_punct(text, keep_math=True)
|
264 |
+
text = text.lower()
|
265 |
+
text = re.sub('\s+',' ',text) # replace multiple space with single
|
266 |
+
return text
|
267 |
+
|
268 |
+
def whisper_norm_text_for_wer(text):
|
269 |
+
# function to format text for wer computation.
|
270 |
+
# uses Whisper normalizer after stripping corpus-specific special tags
|
271 |
+
|
272 |
+
if isinstance(text,list):
|
273 |
+
text = ' '.join(text)
|
274 |
+
text=str(text)
|
275 |
+
text = text.replace('\n',' ') # replace newline with space
|
276 |
+
text = re.sub('%\w+','', text) # remove %HESITATION etc
|
277 |
+
text = remove_in_brackets(text) # removes non-spoken annotations such as [inaudible]
|
278 |
+
normalizer = EnglishTextNormalizer()
|
279 |
+
text = normalizer(text)
|
280 |
+
return text
|
281 |
+
|
282 |
+
def wer_from_df(
|
283 |
+
df,
|
284 |
+
refcol='ref',
|
285 |
+
hypcol='hyp',
|
286 |
+
return_alignments=False,
|
287 |
+
normalise = True,
|
288 |
+
text_norm_method='isat',
|
289 |
+
printout=True):
|
290 |
+
"""Compute WER from a dataframe containing a ref col and a hyp col
|
291 |
+
WER is computed on the edit operation counts over the whole df,
|
292 |
+
not averaged over single utterances.
|
293 |
+
|
294 |
+
Args:
|
295 |
+
df (pandas DataFrame): containing rows per utterance
|
296 |
+
refcol (str, optional): column name containing reference transcript. Defaults to 'ref'.
|
297 |
+
hypcol (str, optional): column name containing hypothesis transcript. Defaults to 'hyp'.
|
298 |
+
return_alignments (bool, optional): Return full word-level alignments. Defaults to False.
|
299 |
+
normalise (bool, optional): Apply text normalisatin to ref and hyp (see norm_text_for_wer). Defaults to True.
|
300 |
+
printout (bool, optional): Print WER metrics. Defaults to True.
|
301 |
+
"""
|
302 |
+
normalizer=get_normalizer(text_norm_method)
|
303 |
+
|
304 |
+
refs=df[refcol].astype(str)
|
305 |
+
hyps = df[hypcol].astype(str)
|
306 |
+
if normalise:
|
307 |
+
refs=refs.apply(normalizer)
|
308 |
+
hyps=hyps.apply(normalizer)
|
309 |
+
|
310 |
+
#ID,ref,hyp,ref_norm,hyp_norm
|
311 |
+
if any(s == '' for s in list(refs)):
|
312 |
+
nonempty=refs.str.len()>0
|
313 |
+
refs=refs[nonempty]
|
314 |
+
hyps=hyps[nonempty]
|
315 |
+
# print(f'{sum(~nonempty)} empty references removed (after normalisation if applied)')
|
316 |
+
wer_meas = jiwer.compute_measures(list(refs), list(hyps))
|
317 |
+
|
318 |
+
if not return_alignments:
|
319 |
+
# remove alignments
|
320 |
+
del wer_meas['ops']
|
321 |
+
del wer_meas['truth']
|
322 |
+
del wer_meas['hypothesis']
|
323 |
+
wer_meas['word_count'] = wer_meas['substitutions']+wer_meas['deletions']+wer_meas['hits']
|
324 |
+
wer_meas['sub_rate'] = wer_meas['substitutions']/wer_meas['word_count']
|
325 |
+
wer_meas['del_rate'] = wer_meas['deletions']/wer_meas['word_count']
|
326 |
+
wer_meas['ins_rate'] = wer_meas['insertions']/wer_meas['word_count']
|
327 |
+
|
328 |
+
if printout:
|
329 |
+
for key in ['wer','sub_rate','del_rate','ins_rate']:
|
330 |
+
print((f"{key}={100*wer_meas[key]:.1f}" ))
|
331 |
+
print(f"word_count={int(wer_meas['word_count'])}")
|
332 |
+
return wer_meas
|
333 |
+
|
334 |
+
|
335 |
+
def wer_from_csv(
|
336 |
+
csv_path,
|
337 |
+
refcol='ref',
|
338 |
+
hypcol='hyp',
|
339 |
+
return_alignments=False,
|
340 |
+
normalise = True,
|
341 |
+
text_norm_method='isat' ,
|
342 |
+
printout=True):
|
343 |
+
|
344 |
+
res = pd.read_csv(csv_path).astype(str)
|
345 |
+
|
346 |
+
wer_meas=wer_from_df(res,
|
347 |
+
refcol=refcol,
|
348 |
+
hypcol=hypcol,
|
349 |
+
return_alignments=return_alignments,
|
350 |
+
normalise = normalise,
|
351 |
+
text_norm_method=text_norm_method,
|
352 |
+
printout=printout)
|
353 |
+
return wer_meas
|
converters.py
ADDED
@@ -0,0 +1,206 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import os
|
2 |
+
import csv
|
3 |
+
import re
|
4 |
+
import pandas as pd
|
5 |
+
from pathlib import Path
|
6 |
+
import numpy as np
|
7 |
+
# functions to convert between different transcript/annotation formats
|
8 |
+
|
9 |
+
#######
|
10 |
+
# "table" refers to a pd.Dataframe w the following cols
|
11 |
+
# [uttID, speaker, transcript, start_sec, end_sec]
|
12 |
+
#########
|
13 |
+
|
14 |
+
# separate function to write to csv, tsv or ELAN compatible (ELAN interprets ALL commas as delimiter so we need to use tab instead)
|
15 |
+
|
16 |
+
def HHMMSS_to_sec(time_str):
|
17 |
+
"""Get Seconds from timestamp string with milliseconds."""
|
18 |
+
if not time_str:
|
19 |
+
return None
|
20 |
+
if time_str.count(':')==2:
|
21 |
+
h, m, s = time_str.split(':')
|
22 |
+
elif time_str.count(':')==3:
|
23 |
+
# weird timestamps where there is a field followign seconds delimited by colon
|
24 |
+
h, m, s, u = time_str.split(':')
|
25 |
+
# determine whether ms field is in tenths or hundredths or thousandths by countng how many digits
|
26 |
+
if len(u)==1:
|
27 |
+
print('Weird time format detected - HH:MM:SS:tenths - please verify this is how you want the time interpreted')
|
28 |
+
ms = float(u)/10
|
29 |
+
elif len(u)==2: # hundredths
|
30 |
+
ms = float(u)/100
|
31 |
+
elif len(u)==3: # hundredths
|
32 |
+
ms = float(u)/1000
|
33 |
+
else:
|
34 |
+
print(f'input string format not supported: {time_str}')
|
35 |
+
return None
|
36 |
+
s = int(s)+ms
|
37 |
+
elif time_str.count(':')==1:
|
38 |
+
# print('missing HH from timestamp, assuming MM:SS')
|
39 |
+
m, s = time_str.split(':')
|
40 |
+
h=0
|
41 |
+
elif time_str.count(':')==0 and time_str.count('.')==1:
|
42 |
+
# print('missing HH:MM from timestamp, assuming SS.ms')
|
43 |
+
s = float(time_str)
|
44 |
+
h=0
|
45 |
+
m=0
|
46 |
+
else:
|
47 |
+
print(f'input string format not supported: {time_str}')
|
48 |
+
return None
|
49 |
+
return int(h) * 3600 + int(m) * 60 + float(s)
|
50 |
+
|
51 |
+
def sec_to_timecode(time_sec):
|
52 |
+
# convert seconds to HH:MM:SS:hundredths as used in .xlsx transcripts
|
53 |
+
h=int(time_sec//3600)
|
54 |
+
m=int((time_sec-3600*h)//60)
|
55 |
+
s=int(time_sec-3600*h-60*m)
|
56 |
+
u=round(100*(time_sec-3600*h-60*m-s))
|
57 |
+
timecode=f'{h}:{m:02}:{s:02}:{u:02}'
|
58 |
+
return(timecode)
|
59 |
+
|
60 |
+
def docx_scraped_tsv_to_table(ooona_file):
|
61 |
+
# ooona output is a table in a word docx,
|
62 |
+
# for now manually copying this out and saving as tsv
|
63 |
+
# but the timestamp format is wrong
|
64 |
+
# input cols are SHOT START END SPEAKER DIALOGUE
|
65 |
+
|
66 |
+
with open(ooona_file) as in_file:
|
67 |
+
reader = csv.reader(in_file, delimiter="\t")
|
68 |
+
next(reader) # skip header
|
69 |
+
rows=[]
|
70 |
+
for i,line in enumerate(reader):
|
71 |
+
utt_ix, start_time, end_time, speaker, transcript = line
|
72 |
+
start_sec = HHMMSS_to_sec(start_time)
|
73 |
+
end_sec = HHMMSS_to_sec(end_time)
|
74 |
+
rows.append([utt_ix,speaker,transcript,start_sec,end_sec])
|
75 |
+
utt_table = pd.DataFrame(rows, columns=['uttID','speaker','transcript','start_sec','end_sec'])
|
76 |
+
return(utt_table)
|
77 |
+
# table = pd.read_csv(ooona_file, sep='\t')
|
78 |
+
|
79 |
+
def molly_xlsx_to_table(xl_file):
|
80 |
+
# contractor transcribers provide an xlsx with the following columns
|
81 |
+
# utt_ix: int
|
82 |
+
# Timecode: "HH:MM:SS:ss - HH:MM:SS:ss"
|
83 |
+
# Duration: HH:MM:SS:ss
|
84 |
+
# Speaker: str
|
85 |
+
# Dialogue: str
|
86 |
+
# Annotations: blank
|
87 |
+
# Error Type: blank
|
88 |
+
with pd.ExcelFile(xl_file) as xls:
|
89 |
+
sheetname = xls.sheet_names
|
90 |
+
table = pd.DataFrame(pd.read_excel(xls, sheetname[0]))
|
91 |
+
table.columns=table.columns.str.lower()
|
92 |
+
table[['start_time','end_time']] = table['timecode'].str.split('-',expand=True)
|
93 |
+
table['start_sec'] = table['start_time'].str.strip().apply(HHMMSS_to_sec)
|
94 |
+
table['end_sec'] = table['end_time'].str.strip().apply(HHMMSS_to_sec)
|
95 |
+
table.drop(labels=['annotations','error type','duration'], axis=1, inplace=True)
|
96 |
+
table=table[['#','speaker','dialogue','start_sec','end_sec']]
|
97 |
+
table.rename(columns={'#':'uttID', 'dialogue':'transcript'}, inplace=True)
|
98 |
+
table.reset_index(inplace=True,drop=True)
|
99 |
+
table=table.replace('', np.nan).dropna(subset=['speaker','dialogue'], how='all') # drop rows with missing values in speaker and utterance
|
100 |
+
return table
|
101 |
+
|
102 |
+
def LoFi_xlsx_to_table(xl_file):
|
103 |
+
# LoFi transcripts have the following columns:
|
104 |
+
# # utt_ix: int
|
105 |
+
# Timecode: "HH:MM:SS:ss - HH:MM:SS:ss"
|
106 |
+
# Duration: HH:MM:SS:ss
|
107 |
+
# Speaker: str
|
108 |
+
# Dialogue: str
|
109 |
+
# Annotations: blank
|
110 |
+
# Error Type: blank
|
111 |
+
with pd.ExcelFile(xl_file) as xls:
|
112 |
+
sheetname = xls.sheet_names
|
113 |
+
table = pd.DataFrame(pd.read_excel(xls, sheetname[0]))
|
114 |
+
table[['start_time','end_time']] = table['Timecode'].str.split('-',expand=True)
|
115 |
+
table['start_sec'] = table['start_time'].str.strip().apply(HHMMSS_to_sec)
|
116 |
+
table['end_sec'] = table['end_time'].str.strip().apply(HHMMSS_to_sec)
|
117 |
+
table.drop(labels=['Annotations','Error Type','Duration'], axis=1, inplace=True)
|
118 |
+
table=table[['#','Speaker','Dialogue','start_sec','end_sec']]
|
119 |
+
table.rename(columns={'#':'uttID','Speaker':'speaker', 'Dialogue':'transcript'}, inplace=True)
|
120 |
+
|
121 |
+
return table
|
122 |
+
|
123 |
+
def saga_to_table(saga_txt):
|
124 |
+
# saga's own transcripts are txt given in the following format
|
125 |
+
#
|
126 |
+
# speaker (start time MM:SS)
|
127 |
+
# utterance
|
128 |
+
# <blank line>
|
129 |
+
# TODO: make more robust by pattern matching instead of modulo
|
130 |
+
with open(saga_txt) as in_file:
|
131 |
+
reader = csv.reader(in_file, delimiter="\n")
|
132 |
+
count = 0
|
133 |
+
rows=[]
|
134 |
+
for i,line in enumerate(reader):
|
135 |
+
print((count,line))
|
136 |
+
if count%3 == 0:
|
137 |
+
# utt = utt.split('\n') # now speaker (time) , transcript
|
138 |
+
# transcript = utt[1]
|
139 |
+
spk_time = line[0].split('(')
|
140 |
+
if len(spk_time)<2:
|
141 |
+
# print('!!!speaker not changed')
|
142 |
+
# print(line)
|
143 |
+
timestamp = spk_time[0].strip('):( ')
|
144 |
+
speaker=rows[-1][0] # prev speaker
|
145 |
+
|
146 |
+
else:
|
147 |
+
speaker = spk_time[0]
|
148 |
+
timestamp = spk_time[1].replace('):','')
|
149 |
+
# print(timestamp)
|
150 |
+
start_sec = HHMMSS_to_sec(timestamp)
|
151 |
+
|
152 |
+
if count%3 == 1:
|
153 |
+
transcript = line[0]
|
154 |
+
if count%3 == 2:
|
155 |
+
rows.append([i,speaker,transcript,start_sec,None])
|
156 |
+
#print([speaker,transcript,timestamp])
|
157 |
+
count+=1
|
158 |
+
utt_table = pd.DataFrame(rows, columns=['uttID','speaker','transcript','start_sec','end_sec'])
|
159 |
+
return(utt_table)
|
160 |
+
|
161 |
+
def table_to_ELAN_tsv(table:pd.DataFrame, path:str):
|
162 |
+
# write table to tsv compatible with ELAN import
|
163 |
+
table.to_csv(path, index=False, float_format='%.3f',sep='\t')
|
164 |
+
|
165 |
+
def table_to_standard_csv(table:pd.DataFrame, path:str):
|
166 |
+
# write table to standard csv format agreed upon by whole team
|
167 |
+
|
168 |
+
# TODO: convert times in seconds back to HH:MM:SS?
|
169 |
+
# TODO: split utterances into sentences?
|
170 |
+
table.to_csv(path,index=False, float_format='%.3f')
|
171 |
+
|
172 |
+
def table_to_utt_labels_csv(table:pd.DataFrame, path:str):
|
173 |
+
# write table to utt_labels csv format comaptable w rosy's isatasr lib
|
174 |
+
table.rename(columns={'transcript':'utterance', 'uttID':'seg'}, inplace=True)
|
175 |
+
table=table.replace('', np.nan).dropna(subset=['speaker','utterance'], how='all') # drop rows with missing values in speaker and utterance
|
176 |
+
table.to_csv(path,index=False, float_format='%.3f')
|
177 |
+
|
178 |
+
def table_to_molly_xlsx(tbl:pd.DataFrame,path:str):
|
179 |
+
tblx = tbl
|
180 |
+
tblx.rename(columns={'uttID':'#', 'speaker':'Speaker','transcript':'Dialogue'}, inplace=True)
|
181 |
+
tblx['dur_s'] = tblx['end_sec']-tblx['start_sec']
|
182 |
+
tblx['start_timecode']=tblx['start_sec'].apply(sec_to_timecode)
|
183 |
+
tblx['end_timecode']=tblx['end_sec'].apply(sec_to_timecode)
|
184 |
+
tblx['Duration'] = tblx['dur_s'].apply(sec_to_timecode)
|
185 |
+
tblx['Timecode'] = [' - '.join(i) for i in zip(tblx['start_timecode'], tblx['end_timecode'])]
|
186 |
+
tblx['Annotations'] = ''
|
187 |
+
tblx['Error Type'] = ''
|
188 |
+
tblx=tblx[['#','Timecode','Duration','Speaker','Dialogue','Annotations','Error Type']]
|
189 |
+
tblx.to_excel(path,sheet_name=Path(path).stem, index=False)
|
190 |
+
|
191 |
+
def utt_labels_csv_to_table(label_csv:str):
|
192 |
+
# utt_labels_csv is the usual format used for diarized, timed transcripts in this repo
|
193 |
+
# There are several versions with differnt columns (with/without segment &/ utterance index)
|
194 |
+
# table:
|
195 |
+
# [uttID, speaker, transcript, start_sec, end_sec]
|
196 |
+
|
197 |
+
table = pd.read_csv(label_csv,keep_default_na=False)
|
198 |
+
# choose which column to use for uttID in table
|
199 |
+
if 'utt' in table.columns:
|
200 |
+
table=table.rename(columns={"utt":"uttID"}).drop('seg', axis=1)
|
201 |
+
elif 'seg' in table.columns:
|
202 |
+
table=table.rename(columns={"seg":"uttID"})
|
203 |
+
else:
|
204 |
+
table=table.reset_index().rename(columns={"index":"uttID"})
|
205 |
+
|
206 |
+
return table
|
renamers.py
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import csv
|
2 |
+
import os
|
3 |
+
import glob
|
4 |
+
import shutil
|
5 |
+
import re
|
6 |
+
|
7 |
+
|
8 |
+
# rename files from original filename (hexadecimal salad) to Session_ID (human readable) and back
|
9 |
+
global DEFAULT_MAP_PATH
|
10 |
+
DEFAULT_MAP_PATH = '../../SessionIDs_from_catalog.csv'
|
11 |
+
|
12 |
+
def make_SessionID_map(path=DEFAULT_MAP_PATH):
|
13 |
+
"""generate dictionary from csv file with columns for File_Name and Session_ID -
|
14 |
+
copied from columsn 1 & 2 of the Catalog on OneDrive
|
15 |
+
"""
|
16 |
+
SID_to_FN={}
|
17 |
+
FN_to_SID={}
|
18 |
+
with open(path,encoding='utf-8-sig') as f:
|
19 |
+
reader = csv.reader(f)
|
20 |
+
headers = next(reader)
|
21 |
+
assert (headers[0]=='File_Name' or headers[0]=='Conference_ID') & (headers[1]=='Session_ID'), "Headers are wrong, expected ('File_Name' or 'Conference_ID') and 'Session_ID'"
|
22 |
+
|
23 |
+
for line in reader:
|
24 |
+
filename,sessionID=line
|
25 |
+
filename=filename.split('.')[0] # remove extensions
|
26 |
+
if (len(filename.strip())>0 and len(sessionID.strip())>0):
|
27 |
+
SID_to_FN[sessionID]=filename
|
28 |
+
FN_to_SID[filename]=sessionID
|
29 |
+
return(SID_to_FN, FN_to_SID)
|
30 |
+
|
31 |
+
|
32 |
+
def rename_files_SID_to_FN(path, recursive=True, overwrite=False):
|
33 |
+
SID_to_FN, _=make_SessionID_map()
|
34 |
+
#TODO: deal with matching nested sIDs, see commented code below
|
35 |
+
newpaths=[]
|
36 |
+
for sID in SID_to_FN.keys():
|
37 |
+
srclist = glob.glob(os.path.join(path,'**', f'*{sID}.*'), recursive=recursive)
|
38 |
+
# print(f'siD: {sID}')
|
39 |
+
# print(srclist)
|
40 |
+
for srcpath in srclist:
|
41 |
+
newpath = srcpath.replace(sID, SID_to_FN[sID])
|
42 |
+
print(newpath)
|
43 |
+
if overwrite==True:
|
44 |
+
shutil.move(srcpath, newpath)
|
45 |
+
else:
|
46 |
+
shutil.copy(srcpath, newpath)
|
47 |
+
newpaths.append(newpath)
|
48 |
+
return newpaths
|
49 |
+
|
50 |
+
|
51 |
+
# # get sessnames
|
52 |
+
# sesslist = [s for s in os.listdir(path) ]
|
53 |
+
# srclist = [os.path.join(src_dir, filename) for filename in os.listdir(src_dir) if os.path.isfile(os.path.join(src_dir, filename))]
|
54 |
+
# for src in srclist:
|
55 |
+
# sessname_matches = [sessname in src for sessname in sesslist]
|
56 |
+
# if sum(sessname_matches)>1:
|
57 |
+
# print('!!!! multiple matches, will take longest match. TODO: implement this you dope')
|
58 |
+
# elif not any(sessname_matches):
|
59 |
+
# print(f'!!!! no sessname matches for file {src}')
|
60 |
+
# else:
|
61 |
+
# sessname = sesslist[sessname_matches.index(True)]
|
62 |
+
# print(f'...copying to {sessname}')
|
63 |
+
# shutil.copy(src, os.path.join(dest_dir,sessname))
|
64 |
+
|
65 |
+
def rename_files_FN_to_SID(path, recursive=True):
|
66 |
+
_, FN_to_SID=make_SessionID_map()
|
67 |
+
|
68 |
+
def extract_conferenceID_from_filename(filename):
|
69 |
+
"""extract conferenceID from filename
|
70 |
+
"""
|
71 |
+
conferenceID=filename.split(' ')[0]
|
72 |
+
conferenceID = re.sub('_?[a-zA-Z]*(\.*[a-zA-Z]*).xlsx','', conferenceID)
|
73 |
+
conferenceID=re.sub('TMcoded|Transcript','', conferenceID)
|
74 |
+
conferenceID=re.sub('_start\d+_end\d+_?','', conferenceID)
|
75 |
+
conferenceID=re.sub(
|
76 |
+
'\d{5}_\d{4}-\d{2}-\d{2}_','', conferenceID)
|
77 |
+
return conferenceID
|
trimmers.py
ADDED
@@ -0,0 +1,139 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from pathlib import Path
|
2 |
+
import os
|
3 |
+
import csv
|
4 |
+
import subprocess
|
5 |
+
import pandas as pd
|
6 |
+
import sys
|
7 |
+
sys.path.append('..')
|
8 |
+
|
9 |
+
from levi.converters import HHMMSS_to_sec
|
10 |
+
|
11 |
+
def trim_media(media_in,
|
12 |
+
media_out,
|
13 |
+
start,
|
14 |
+
end):
|
15 |
+
|
16 |
+
# options for writing out audio if converting
|
17 |
+
WAV_CHANNELS = 1
|
18 |
+
WAV_SAMPLE_RATE = 16000
|
19 |
+
|
20 |
+
media_type = Path(media_in).suffix
|
21 |
+
ext = Path(media_out).suffix
|
22 |
+
|
23 |
+
if isinstance(start, str):
|
24 |
+
start_sec = HHMMSS_to_sec(start)
|
25 |
+
else:
|
26 |
+
start_sec = float(start)
|
27 |
+
if isinstance(end, str):
|
28 |
+
end_sec = HHMMSS_to_sec(end)
|
29 |
+
else:
|
30 |
+
end_sec = float(end)
|
31 |
+
|
32 |
+
if ext == '.wav':
|
33 |
+
# convert to wav with standard format for audio models
|
34 |
+
print(f'...Using ffmpeg to trim video from {start} to {end} \n and convert to {WAV_SAMPLE_RATE}Hz WAV with {WAV_CHANNELS} channels...')
|
35 |
+
print(f'...generating {media_out}...')
|
36 |
+
|
37 |
+
subprocess.call(['ffmpeg',
|
38 |
+
'-y',
|
39 |
+
'-i',
|
40 |
+
media_in,
|
41 |
+
'-ss',
|
42 |
+
f'{start_sec}',
|
43 |
+
'-to',
|
44 |
+
f'{end_sec}',
|
45 |
+
'-acodec',
|
46 |
+
'pcm_s16le',
|
47 |
+
'-ac',
|
48 |
+
WAV_CHANNELS,
|
49 |
+
'-ar',
|
50 |
+
WAV_SAMPLE_RATE,
|
51 |
+
media_out,
|
52 |
+
'-hide_banner',
|
53 |
+
'-loglevel',
|
54 |
+
'warning'
|
55 |
+
],shell=False)
|
56 |
+
|
57 |
+
else:
|
58 |
+
|
59 |
+
print(f'...Using ffmpeg to trim video from {start_sec} to {end_sec}...')
|
60 |
+
print(f'...generating {media_out}...')
|
61 |
+
|
62 |
+
subprocess.call(['ffmpeg',
|
63 |
+
'-y',
|
64 |
+
'-i',
|
65 |
+
media_in,
|
66 |
+
'-ss',
|
67 |
+
f'{start_sec}',
|
68 |
+
'-to',
|
69 |
+
f'{end_sec}',
|
70 |
+
'-c',
|
71 |
+
'copy',
|
72 |
+
media_out,
|
73 |
+
'-hide_banner',
|
74 |
+
'-loglevel',
|
75 |
+
'warning'
|
76 |
+
],shell=False)
|
77 |
+
|
78 |
+
def trim_media_batch(extract_timings_csv,
|
79 |
+
outpath,
|
80 |
+
suffix='',
|
81 |
+
convert_to=False):
|
82 |
+
"""trim a batch of media files given a csv of timings
|
83 |
+
|
84 |
+
Args:
|
85 |
+
extract_timings_csv (str): path to csv with columns:
|
86 |
+
filepath, start (HH:MM:SS), end (HH:MM:SS)
|
87 |
+
outpath (str): output path
|
88 |
+
suffix (str, optional): save output trimmed files with this suffix. Defaults to ''.
|
89 |
+
convert_to (bool, optional): [None, 'wav','mp4']. Defaults to False.
|
90 |
+
Returns:
|
91 |
+
outfiles (list): list of file paths created
|
92 |
+
"""
|
93 |
+
|
94 |
+
|
95 |
+
|
96 |
+
os.makedirs(outpath, exist_ok=True)
|
97 |
+
|
98 |
+
samples_df = pd.read_csv(
|
99 |
+
extract_timings_csv,
|
100 |
+
skip_blank_lines=True,
|
101 |
+
index_col=False,
|
102 |
+
names=['media_in','startHMS','endHMS'],
|
103 |
+
header=0
|
104 |
+
).dropna().sort_values(
|
105 |
+
by='media_in',ignore_index=True).reset_index(drop=True)
|
106 |
+
|
107 |
+
print(f'TRIMMING {len(samples_df.index)} FILES...')
|
108 |
+
|
109 |
+
# enumerate samples by session and check if there are multiple samples from a given session
|
110 |
+
samples_df['count'] = samples_df.groupby('media_in').cumcount()
|
111 |
+
if not os.path.exists(outpath):
|
112 |
+
os.makedirs(outpath)
|
113 |
+
|
114 |
+
outfiles=[]
|
115 |
+
for i, rec in samples_df.iterrows():
|
116 |
+
media_in,startHMS,endHMS, count = rec.values
|
117 |
+
suffix_use = f'{suffix}{count}' if count > 0 else suffix # if multiple samples per recording, give a diffrent name
|
118 |
+
|
119 |
+
if not os.path.exists(media_in):
|
120 |
+
print(f'!!!WARNING: media not found: {media_in}')
|
121 |
+
continue
|
122 |
+
|
123 |
+
media_type = Path(media_in).suffix
|
124 |
+
sessname = Path(media_in).stem
|
125 |
+
print(f'...Input media: {media_in}')
|
126 |
+
|
127 |
+
if convert_to=='wav':
|
128 |
+
ext = '.wav'
|
129 |
+
elif convert_to=='mp4':
|
130 |
+
ext = '.mp4'
|
131 |
+
else:
|
132 |
+
ext = media_type
|
133 |
+
|
134 |
+
outfile = os.path.expanduser(os.path.join(outpath,f'{sessname}{suffix_use}{ext}'))
|
135 |
+
|
136 |
+
trim_media(media_in, outfile, HHMMSS_to_sec(startHMS), HHMMSS_to_sec(endHMS))
|
137 |
+
|
138 |
+
outfiles.append(outfile)
|
139 |
+
return(outfiles)
|