Making Transcripts with WhisperX

1. Enter the following code into the terminal: pip install whisperx.

2. In that same respository, add a folder of your source files.

3. Enter the following prompt into the terminal:

whisperx {filepath} --model turbo --output_dir output

You may need more commands depending on what you want from WhisperX.

Special prompts:

If the audio is in a language other than English, append the following code to your command prompt:

--language {Language_Name}

To add diarization, append the following code:

--diarize

To help diarization, append the following code:

--min_speakers {Number} --max_speakers {Number}

Keep in mind that commercials, announcements, and narrators are also considered speakers.

WhisperX output

As output, five files will be generated: .json, .srt, .tsv, .txt, and .vtt

To move the json and vtt to whisper-editor, make sure you have the following add-to-json.py downloaded.

from csv import DictReader
import os
import json
import shutil
json_file_data = "/Users/{Username}/whisper-reviewer/public/test_for_whisper_transcript_online.json"
all_data = []

with open(json_file_data, "r") as my_json:
audio_file_data = json.load(my_json)
with open("mycsv.csv", "r") as my_file:
reader = DictReader(my_file)
for row in reader:
    if row['Title'] != "":
        all_data.append(
            {
                "uin": row['UIN'],
                "title": row['Title'],
                "audio": row['Audio']
            }
        )
for path, directories, files in os.walk("output"):
    for file in files:
        if ".json" in file:
            identifier = file.split('.json')[0]
            name = [item["title"] for item in all_data if item['uin'] == identifier][0]
            full_name = f"{name} whisperx"
            audio_file_location = [item["audio"] for item in all_data if item['uin'] == identifier][0]
            audio_file_data.append(
                {
                    "audio": audio_file_location,
                    "url": f"./transcripts/{file}",
                    "name": full_name
                }
            )
            shutil.move(f"{path}/{file}", "/Users/{Username}/whisper-reviewer/public/transcripts")
            with open(json_file_data, "w") as f:
                json.dump(audio_file_data, f, indent=4)
        elif '.vtt' in file:
            shutil.move(f"{path}/{file}", "/Users/{Username}/whisper-reviewer/vtts/{output-folder}")