pythonr writer optiions not respected

[Linch1]

Hello, i'm trying to create the same result as the terminal command

whisper path.wav --word_timestamps True --max_line_width 22 --max_line_count 1 --highlight_words True

but i'm not getting the same result. I think that the code is almost complete but i'm not getting what option is missing or why it generates different results

import whisper
from whisper.utils import get_writer

def whisper_transcription(input_audio, whisper_model):

    model = whisper.load_model(whisper_model)
    result = model.transcribe(input_audio, verbose=True)
    output_directory = "."

    # Set some initial options values
    options = {
        'max_line_width': 22,
        'max_line_count': 1,
        'highlight_words': True
    }

   # Include options in call
    txt_writer = get_writer("txt", output_directory)
    txt_writer(result, input_audio, options)

    json_writer = get_writer("json",  output_directory)
    json_writer(result, input_audio, options)

    vtt_writer = get_writer("vtt", output_directory)
    vtt_writer(result, input_audio, options)

    tsv_writer = get_writer("tsv", output_directory)
    tsv_writer(result, input_audio, options)

    srt_writer = get_writer("srt", output_directory)
    srt_writer(result, input_audio, options)

audio_path = "path.wav"
whisper_transcription(audio_path, 'base')

What am i missing?

Python output srt

1
00:00:00,000 --> 00:00:07,120
There's a really embarrassing moment. I definitely erased from my life. Once I arranged an underwater

Command line output srt

1
00:00:00,000 --> 00:00:00,360
Um, there's a really

2
00:00:00,360 --> 00:00:00,540
Um, there's a really

3
00:00:00,540 --> 00:00:00,900
Um, there's a really

.. and so on, avoinding posting all the result since it useless.

[phineas-pta]

enable word_timestamps