Script adapted from: https://github.com/m-bain/whisperX

Create a virtual environment using requirements.txt

'''

Prepare and save transcription & speaker diarization from an audio file.

Parameters
----------
audio_file: Path
    path & audio file name to process, .wav
output_path: Path
    path to save output files
max_speakers : integer
    number of speakers in the conversation, default = 2

Returns
-------
audio file name + transcription.txt - audio transcription
audio file name + diarization.json - speaker diarization
audio file name + combined_result_json.json - transcription & diarization combined, in .json format
audio file name + combined_result_tsv.tsv - transcription & diarization combined, in .tsv format

command line: $ python x_asr.py audio_file output_path max_speakers

'''