split audio_segmentation with corresponding transcript for youtube datasets
first downloading .mp3 playlist for youbute indian speakers with .vtt subtitle file.
.vtt file format like starting-ending timing with the audio transcript. i was segmenting that youtube audiofile with Start-End time.
and i applied some preprocessing like data cleaning, wav file format 16bit 16khz mono, and then use it deepspeech training.
youtube_news.txt
python3 youtube_download.py
python3 text1.py