yt_transcriber

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# Filename: yt_transcriber.py
# Version: 1.0.1
# Author: Jeoi Reqi

"""
This script is a utility for extracting transcripts from YouTube shorts & videos.
It provides functions to extract the video ID from a YouTube URL, fetch the transcript for the specified video, and save the transcript to a text file.
The script also includes a user-friendly interface for selecting options to input a YouTube URL or use an example URL, and it supports fetching transcripts in multiple languages.

Requirements:
    - youtube_transcript_api: This script requires the 'youtube_transcript_api' library.
    - Install via pip command:

        'pip install youtube_transcript_api'

Functions:
    1. extract_video_id(video_url): Extracts the video ID from a YouTube URL.
    2. save_transcript_to_file(transcript_text): Saves the transcript text to a text file.
    3. main(): Provides a user interface for selecting options to input a YouTube URL or use an example URL, fetches the transcript, and offers the option to save the transcript to a file.

Usage:
    - Run the script using Python: `python yt_transcriber.py`
    - Select an option:
        1. Input YouTube Video Url: Enter a YouTube video or Shorts URL to fetch the transcript.
        2. Use Example Video Url: Use an example video URL provided in the script to fetch the transcript.
    - If selecting option 1:
        - Enter the YouTube Video or Shorts URL.
        - The script attempts to fetch the transcript in specified languages (English, Spanish, Arabic, Simplified Chinese).
        - The transcript is displayed, and the user is prompted to save it to a file.
    - If selecting option 2:
        - The transcript for the example video URL is fetched and displayed.
        - The user is prompted to save the transcript to a file.

Additional Notes:
    - This script supports fetching transcripts in multiple languages, including English, Spanish, Arabic, and Simplified Chinese.
    - The script handles various user inputs and provides appropriate error messages for invalid inputs.
    - Make sure to install the 'youtube_transcript_api' library before running the script.
"""

from youtube_transcript_api import YouTubeTranscriptApi

def extract_video_id(video_url):
    """
    Extracts the video ID from a YouTube URL.

    Args:
        video_url (str): The YouTube video URL.

    Returns:
        str or None: The extracted video ID if found, otherwise None.
    """
    # Extract video ID from YouTube URL
    if "youtube.com" in video_url:
        if "/shorts/" in video_url:
            video_id = video_url.split("/shorts/")[-1]
            return video_id
        else:
            video_id = video_url.split("v=")[-1]
            return video_id
    elif "youtu.be" in video_url:
        video_id = video_url.split("/")[-1]
        return video_id
    else:
        return None

def save_transcript_to_file(transcript_text):
    """
    Saves the transcript text to a text file.

    Args:
        transcript_text (str): The transcript text to be saved.
    """
    file_name = input(
        "\nEnter the file name to save the transcript (without extension): "
    )
    file_path = f"{file_name}.txt"
    with open(file_path, "w", encoding="utf-8") as file:
        file.write(transcript_text)
    print(f"\nTranscript saved as {file_path}")

def main():
    # Print The Header
    print(
        r"""
      __   __         _____      _
      \ \ / /__  _   |_   _|   _| |__   ___
       \ V / _ \| | | || || | | | '_ \ / _ \
        | | (_) | |_| || || |_| | |_) |  __/
 _____  |_|\___/ \__,_||_| \__,_|_.__/_\___|
|_   _| __ __ _ _ __  ___  ___ _ __(_) |__   ___ _ __
  | || '__/ _` | '_ \/ __|/ __| '__| | '_ \ / _ \ '__|
  | || | | (_| | | | \__ \ (__| |  | | |_) |  __/ |
  |_||_|  \__,_|_| |_|___/\___|_|  |_|_.__/ \___|_|
        """
    )

    # Main Menu
    print("\n\tSelect an option:\n")
    print("\t1: Input YouTube Video or Shorts Url")
    print("\t2: Use Example Video Url")
    option = input("\n\tEnter the option number (1 or 2): ")

    if option == "1":
        video_url = input("\n\tEnter the YouTube Video URL: ")
        video_id = extract_video_id(video_url)

        if not video_id:
            print("\nInvalid YouTube Video URL. Exiting...")
            return

        # Scan CC for the specified languages
        languages_to_try = ["en", "es", "ar", "zh-Hans"]
        transcript = None

        # Attempt to fetch transcripts in specified languages
        for lang in languages_to_try:
            transcript = YouTubeTranscriptApi.get_transcript(video_id, languages=[lang])
            if transcript:
                break

        if not transcript:
            print("\nNo transcript available for the selected languages. Exiting...")
            return

        transcript_text = ""
        for item in transcript:
            transcript_text += f"{item['start']:>6.2f}: {item['text']}\n"
        print(transcript_text)

        # Save Menu
        print("\n\tDo you want to save the transcript to a file?\n")
        print("\t1: Yes")
        print("\t2: No")
        save_option = input("\nEnter the option number: ")

        # Function to save the transcript to a txt file in the current working directory
        if save_option == "1":
            save_transcript_to_file(transcript_text)
        elif save_option == "2":
            print("\nTranscript not saved!")
        else:
            print("\nInvalid option!")
    elif option == "2":
        video_id = "jNQXAC9IVRw"  # Example video URL: https://www.youtube.com/watch?v=jNQXAC9IVRw
        transcript = YouTubeTranscriptApi.get_transcript(video_id)
        transcript_text = ""
        for item in transcript:
            transcript_text += f"{item['start']:>6.2f}: {item['text']}\n"
        print(transcript_text)

        # Save Menu
        print("\n\tDo you want to save the transcript to a file?\n")
        print("\t1: Yes")
        print("\t2: No")
        save_option = input("\n\tEnter the option number: ")

        # Function to save the transcript to a txt file in the current working directory
        if save_option == "1":
            save_transcript_to_file(transcript_text)
        elif save_option == "2":
            print("\nTranscript not saved!")
        else:
            print("\nInvalid option!")
    else:
        print("\nInvalid option!")

if __name__ == "__main__":
    main()