A couple of helper scripts to generate a .srt file for a given .wav file input.
Uses OpenAI's Whisper and the faster-whisper implementation.
- Install the necessary dependencies.
- Ensure you have a valid .wav audio file for transcription.
- Run the script with the desired command-line options and provide the path to the input audio file.
- The script will transcribe the audio and generate a corresponding .srt subtitle file.
python main_faster_whisper.py --input audio.wav --language en --remove-silence
This will start transcribing the file audio.wav while assuming it's in English. Additionally, it will remove silent portions from the audio file longer than 2 seconds, which helps to get a better transcription. The resulting transcription will be written to audio.srt.
To get more info, refer to the script's --help option:
python main_faster_whisper.py --help
A tool to easily shift timestamps of an .srt file!
The script takes an input file (input_file) which should be an .srt file containing the subtitles you want to shift. The shifted subtitles will be written to the output file (output_file). The shift_amount argument specifies the amount of seconds by which the subtitles should be shifted.
python subtitles_shift.py --input_file input.srt --output_file output.srt --shift_amount 5
This will write to the file output.srt where all the timestamps from input.srt will be shifted by 5 seconds.