Table of Contents

• Getting Started
  • Prerequisites
  • Installation
• Usage
• License
• Contact

Getting Started

Prerequisites

Python 3.11 or above

Note for Ubuntu/Debian: This project uses python-magic to identify file types. So the libmagic C library needs to be installed need to be installed. See python-magic for more information.

For Debian/Ubuntu systems:

• sudo apt-get install libmagic1

For OSX:

• When using Homebrew: brew install libmagic
• When using macports: port install file

For Windows:

• No additional requirements

Installation

  pip install ak_transcribe

Check configs

  import ak_transcribe
  ak_transcribe.test_configs()

Usage

from ak_transcribe import Transcriber
from pathlib import Path
res = Transcriber.process(filepath=Path("path/to/media/file"))

# Write Json to file
with open(filepath.with_suffix('.json'), 'w', encoding='utf-8') as f:
    f.write(res.json)

# Return Srt file contents
res.srt

# Embed subtitle file into the video
res.embed_srt(filepath=Path("path/to/media/file"))

# Return Transcript text
res.txt

# Extract txt from srt file
from ak_transcribe import converter
converter.srt_to_txt("path/to/srt/file")

# Read existing srt
from ak_transcribe import TranscribedContent
TranscribedContent(Path("path/to/JSON/file"))

## Alternatively, pass dictionary from json file
import json
with open(Path("path/to/JSON/file"), "r", encoding="utf-8") as f:
    return TranscribedContent(results=json.load(f))

License

See LICENSE for more information.

Contact

Arun Kishore - @rpakishore

Project Link: https://github.com/rpakishore/ak_transcribe