import time import requests import re import gradio as gr import os import shutil api_key = "268976:66f4f58a2a905" def read_srt_file(file_path): try: with open(file_path, 'r', encoding='utf-8') as file: srt_content = file.read() return srt_content except FileNotFoundError: print(f"The file {file_path} was not found.") except Exception as e: print(f"An error occurred: {e}") def clean_text(text): # Remove 'srt ' from the start of each line # Remove ''' from the start and end text = re.sub(r"^```|```$", '', text) text = re.sub(r'^srt', '', text, flags=re.MULTILINE) return text def translate_text(api_key, text, source_language = "en", target_language = "fa"): url = "https://api.one-api.ir/translate/v1/google/" request_body = {"source": source_language, "target": target_language, "text": text} headers = {"one-api-token": api_key, "Content-Type": "application/json"} response = requests.post(url, headers=headers, json=request_body) if response.status_code == 200: result = response.json() return result['result'] else: print(f"Error: {response.status_code}, {response.text}") return None def enhance_text(api_key, text): url = "https://api.one-api.ir/chatbot/v1/gpt4o/" # Prepare the request body request_body = [{ "role": "user", "content": f"Please take the following SRT subtitle text in English and translate only the subtitle text into Persian. Ensure that all numbering and time codes remain unchanged. convert English terms in to common persian terms. The output should be a new SRT file with the subtitles in Persian, preserving the original formatting and timings and exept for the subtitle dont return anything in response. the subtitle will be provided in the following message" }, { "role": "assistant", "content": "okay" }, { "role": "user", "content": text } ] # Add the API key to the request headers = { "one-api-token": api_key, "Content-Type": "application/json" } # Make the POST request attempts = 0 max_attempts = 3 while attempts < max_attempts: response = requests.post(url, headers=headers, json=request_body) if response.status_code == 200: result = response.json() if result["status"] == 200: print("status: ", result["status"]) te = clean_text(result["result"][0]) print("result: ", te) return te else: print(f"Error: status {result['status']}, retrying in 30 seconds...") else: print(f"Error: {response.status_code}, {response.text}, retrying in 30 seconds...") attempts += 1 time.sleep(10) print("Error Max attempts reached. Could not retrieve a successful response.") te = translate_text(api_key, text) return te def generate_translated_subtitle(language, segments, input_video_name): input_video_name=input_video_name.split('/')[-1] subtitle_file = f"{input_video_name}.srt" text = "" lines = segments.split('\n') new_list = [item for item in lines if item != ''] segment_number = 1 for index, segment in enumerate(new_list): if (index+1) % 3 == 1 or (index+1)==1: text += f"{segment}\n" segment_number += 1 if (index+1) % 3 == 2 or (index+1)==2: text += segment + "\n" if (index+1) % 3 == 0: text += f"\u200F{segment}\n\n" with open(subtitle_file, "a", encoding='utf8') as f: f.write(text) return subtitle_file def split_srt_file(input_file, max_chars=3000): # Read the contents of the SRT file with open(input_file, 'r', encoding='utf-8') as file: content = file.read() file.close() # Split the content into individual subtitles subtitles = content.strip().split('\n\n') # Prepare to write the split files output_files = [] current_file_content = '' current_file_index = 1 for subtitle in subtitles: # Check if adding this subtitle would exceed the character limit if len(current_file_content) + len(subtitle) + 2 > max_chars: # +2 for \n\n # Write the current file output_file_name = f'split_{current_file_index}.srt' with open(output_file_name, 'w', encoding='utf-8') as output_file: output_file.write(current_file_content.strip()) output_files.append(output_file_name) # Prepare for the next file current_file_index += 1 current_file_content = subtitle + '\n\n' else: # If it fits, add the subtitle current_file_content += subtitle + '\n\n' # Write any remaining content to a new SRT file if current_file_content: output_file_name = f'split_{current_file_index}.srt' with open(output_file_name, 'w', encoding='utf-8') as output_file: output_file.write(current_file_content.strip()) output_files.append(output_file_name) return output_files def translate(file, max_chars): print("translate") srt_files = split_srt_file(file, max_chars=3000) for srt_file in srt_files: srt = read_srt_file(srt_file) srt_string = enhance_text(api_key, srt) print(srt_string) subtitle_file = generate_translated_subtitle('fa', srt_string, 'video_subtitled') time.sleep(10) if os.path.exists(subtitle_file): copied_subtitle = os.path.join(os.path.dirname(subtitle_file), f"copy_{os.path.basename(subtitle_file)}") shutil.copy2(subtitle_file, copied_subtitle) # Delete the original subtitle file os.remove(subtitle_file) return copied_subtitle else: print("Error: subtitle file not found.") return None with gr.Blocks() as demo: gr.Markdown("Start typing below and then click **Run** to see the progress and final output.") with gr.Column(): srt_file = gr.File() max_chars = gr.Number() subtitle_file = gr.File() btn = gr.Button("Create") btn.click( fn=translate, inputs=[srt_file, max_chars], outputs=subtitle_file, ) demo.launch(debug=True)