Build a script to reformat meeting transcripts

Here’s a Python script that can reformat meeting transcripts by:

• Adding speaker labels clearly

• Adding timestamps if present

• Cleaning up spacing and line breaks

• Organizing the text for easier reading

This script assumes the input transcript is a plain text file, where speaker names and their dialogue are somewhat identifiable. You can customize patterns based on your transcript style.

python
CopyEdit
import re

def reformat_transcript(input_text):
    # Example pattern to identify speaker lines: "SpeakerName: speech"
    # Adjust regex if your transcript format differs
    speaker_pattern = re.compile(r'^(.*?):s*(.*)', re.MULTILINE)

    # Find all matches of speaker and speech
    matches = speaker_pattern.findall(input_text)

    reformatted_lines = []
    for speaker, speech in matches:
        # Clean extra spaces and line breaks in speech
        speech = ' '.join(speech.split())
        reformatted_lines.append(f"{speaker.strip()}:n  {speech}n")

    return 'n'.join(reformatted_lines)

def main():
    input_file = 'transcript.txt'
    output_file = 'reformatted_transcript.txt'

    with open(input_file, 'r', encoding='utf-8') as f:
        transcript = f.read()

    reformatted = reformat_transcript(transcript)

    with open(output_file, 'w', encoding='utf-8') as f:
        f.write(reformatted)

    print(f"Reformatted transcript saved to {output_file}")

if __name__ == '__main__':
    main()

How this works:

• It scans the transcript for lines starting with SpeakerName:.

• Cleans and condenses the speech text.

• Formats each speaker’s line with the speaker on one line and indented speech below.

• Saves the output to a new file.

If your transcript has timestamps or different formats, share a sample, and I can tweak the script accordingly.