Feature request Output text as txt, vtt and srt

[Cate9021]

Thanks for develop this cool whisper UI project, would like to ask if possible to output the result as .txt, .vtt and .srt ?

It's already in whisper/transcribe.py:

` for audio_path in args.pop("audio"): result = transcribe(model, audio_path, temperature=temperature, **args)

    audio_basename = os.path.basename(audio_path)

    # save TXT
    with open(os.path.join(output_dir, audio_basename + ".txt"), "w", encoding="utf-8") as txt:
        write_txt(result["segments"], file=txt)

    # save VTT
    with open(os.path.join(output_dir, audio_basename + ".vtt"), "w", encoding="utf-8") as vtt:
        write_vtt(result["segments"], file=vtt)

    # save SRT
    with open(os.path.join(output_dir, audio_basename + ".srt"), "w", encoding="utf-8") as srt:
        write_srt(result["segments"], file=srt)`

[hayabhay]

Yup! That should be doable! I'll be releasing a big update in the next few days and will add this in :)

[cooper-ganglia]

Any update on this? I didn't see the option available (but then again, I'm a noob and my "Instructions" page is broken due to some operator error, lol). I'd love to be able to export SRT captions straight from this awesome app!

[dorjeduck]

I was just looking for the exactly same feature and as it seems not yet included I just added it to the file transcribe.py as follows (just elaborating on the answer by Cate9021 so that it works with the current version)

on the top of transcribe.py add the following two imports

import os from whisper.utils import write_srt,write_vtt,write_txt

and in the method def transcribe add

self.text = self.raw_output["text"] self.language = self.raw_output["language"] self.segments = self.raw_output["segments"]

    ## -----------new code start

    transcript_basename = self.name + '__' + whisper_model  

    # save TXT
    with open(os.path.join(self.save_dir, transcript_basename + ".txt"), "w", encoding="utf-8") as txt:
        write_txt(self.segments, file=txt)

    # save VTT
    with open(os.path.join(self.save_dir, transcript_basename + ".vtt"), "w", encoding="utf-8") as vtt:
        write_vtt(self.segments, file=vtt)

    # save SRT
    with open(os.path.join(self.save_dir, transcript_basename + ".srt"), "w", encoding="utf-8") as srt:
        write_srt(self.segments, file=srt)

    ## -----------new code end

    # Remove token ids from the output
    for segment in self.segments:
        del segment["tokens"]

This creates the transcript files in the folder local/[Audio/Video Name] when you are transcribing an audio/video. The variable transcript_basename sets the file name which you might want to change according to your preferences (i added the model name as i am evaluating the different models right now)

[hayabhay]

Apologies for the delay. This is now implemented. Each audio file is saved locally and transcriptions on all formats are saved to the same directory (#13)