mirix commited on
Commit
2f5721d
1 Parent(s): 8b7437e

Upload parquet2csv_wav.py

Browse files
Files changed (1) hide show
  1. parquet2csv_wav.py +18 -1
parquet2csv_wav.py CHANGED
@@ -10,29 +10,46 @@ import numpy as np
10
  from scipy.io.wavfile import write
11
 
12
  import os
 
 
13
  n_cores = str(os.cpu_count())
14
  os.environ['OMP_NUM_THREADS'] = n_cores
15
  os.environ['MKL_NUM_THREADS'] = n_cores
16
 
 
17
  if not os.path.isdir('wavs'):
18
  os.makedirs('wavs')
19
 
 
20
  columns = ['ytid', 'ytid_seg', 'start', 'end', 'sentiment', 'happiness', 'sadness', 'anger', 'fear', 'disgust', 'surprise']
21
 
 
 
 
 
 
22
  df = pl.read_parquet('sqe_messai.parquet', columns = columns)
23
 
 
24
  df.write_csv('sqe_messai_nowav.csv')
25
-
26
  print(df)
27
 
 
28
  columns2 = ['ytid_seg', 'wav2numpy']
29
 
 
30
  df2 = pl.read_parquet('sqe_messai.parquet', use_pyarrow=False, columns = columns2)
31
 
 
 
 
 
 
32
  def numpy2wav(row):
33
  segment = os.path.splitext(os.path.basename(os.path.normpath(row[0])))[0]
34
  print('PROCESSED:', segment)
35
  write('wavs/' + segment + '.wav', 16000, np.array(row[1]))
36
  return segment
37
 
 
38
  df2.apply(lambda x: numpy2wav(x))
 
10
  from scipy.io.wavfile import write
11
 
12
  import os
13
+ # User all available cores
14
+ # It seems useless in the context of this script
15
  n_cores = str(os.cpu_count())
16
  os.environ['OMP_NUM_THREADS'] = n_cores
17
  os.environ['MKL_NUM_THREADS'] = n_cores
18
 
19
+ # Create the wavs dir if it does not exist
20
  if not os.path.isdir('wavs'):
21
  os.makedirs('wavs')
22
 
23
+ # All columns from the parquet file except the one with the audio numpy arrays (it is huge)
24
  columns = ['ytid', 'ytid_seg', 'start', 'end', 'sentiment', 'happiness', 'sadness', 'anger', 'fear', 'disgust', 'surprise']
25
 
26
+ ### Add replace line here to change the paths from the ytid_seg column ###
27
+ # https://pola-rs.github.io/polars/py-polars/html/reference/expressions/api/polars.Expr.str.replace.html
28
+ # replace the subtring '/home/emoman/Downloads/mosei/samples' with the actual path
29
+
30
+ # Read the parquet file with polars
31
  df = pl.read_parquet('sqe_messai.parquet', columns = columns)
32
 
33
+ # Export the csv file (excluding the last column)
34
  df.write_csv('sqe_messai_nowav.csv')
 
35
  print(df)
36
 
37
+ # Now we are only interested on the column with the paths and the audio numpy arrays
38
  columns2 = ['ytid_seg', 'wav2numpy']
39
 
40
+ # Read the parquet file with polars (this will take a while)
41
  df2 = pl.read_parquet('sqe_messai.parquet', use_pyarrow=False, columns = columns2)
42
 
43
+ ### Add replace line here to change the paths from the ytid_seg column ###
44
+ # https://pola-rs.github.io/polars/py-polars/html/reference/expressions/api/polars.Expr.str.replace.html
45
+ # replace the subtring '/home/emoman/Downloads/mosei/samples' with the actual path
46
+
47
+ # Function to convert the numpy arrays to wav files stored in the wavs folders
48
  def numpy2wav(row):
49
  segment = os.path.splitext(os.path.basename(os.path.normpath(row[0])))[0]
50
  print('PROCESSED:', segment)
51
  write('wavs/' + segment + '.wav', 16000, np.array(row[1]))
52
  return segment
53
 
54
+ # Apply the function (this will take a while)
55
  df2.apply(lambda x: numpy2wav(x))