experiments/STT
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Merge pull request #3380 from piraka9011/patch-1

Browse Source 
Convert channels for CV2 dataset
This commit is contained in:
Reuben Morais 2020-10-17 00:43:08 +02:00 committed by GitHub
commit 0798698e97
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
bin/import_cv2.py
View File

@ -27,6 +27,7 @@ from ds_ctcdecoder import Alphabet
FIELDNAMES = ["wav_filename", "wav_filesize", "transcript"] FIELDNAMES = ["wav_filename", "wav_filesize", "transcript"]
SAMPLE_RATE = 16000 SAMPLE_RATE = 16000
CHANNELS = 1
MAX_SECS = 10 MAX_SECS = 10
PARAMS = None PARAMS = None
FILTER_OBJ = None FILTER_OBJ = None
@ -179,7 +180,7 @@ def _preprocess_data(tsv_dir, audio_dir, space_after_every_character=False):
def _maybe_convert_wav(mp3_filename, wav_filename): def _maybe_convert_wav(mp3_filename, wav_filename):
if not os.path.exists(wav_filename): if not os.path.exists(wav_filename):
transformer = sox.Transformer() transformer = sox.Transformer()
transformer.convert(samplerate=SAMPLE_RATE) transformer.convert(samplerate=SAMPLE_RATE, n_channels=CHANNELS)
try: try:
transformer.build(mp3_filename, wav_filename) transformer.build(mp3_filename, wav_filename)
except sox.core.SoxError: except sox.core.SoxError: