Keep .vtt as string object vs write to file - OpenAI Whisper

Question

I am using OpenAI's whisper to take an .mp4 file and convert it to a .vtt file. This works well for the most part, here is an example:

my_file = "filepath.mp4"
model = whisper.load_model("tiny")
result = model.transcribe(my_file )
vtt_writer = whisper.utils.WriteVTT(output_dir="./")
vtt_writer(result, my_file)

This produces a .vtt file for me that looks all correct.

However, I want to, instead of writing to a file, keep it as a string variable.

Simply assigning vtt=vtt_writer(result, my_file) doesn't result in anything (i.e. vtt is an empty object).

I was able to find where this is happening;

class WriteVTT(SubtitlesWriter):
    extension: str = "vtt"
    always_include_hours: bool = False
    decimal_marker: str = "."

    def write_result(self, result: dict, file: TextIO, options: dict):
        print("WEBVTT\n", file=file)
        for start, end, text in self.iterate_result(result, options):
            print(f"{start} --> {end}\n{text}\n", file=file, flush=True)

However, I'd like to avoid:

• Reworking/editing the source code if I can
• Writing to file, reading it back in, then deleting it from file

What other options do I have?