Spaces:
Running
Running
File size: 3,522 Bytes
9c1d41d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 |
import pysrt
from openai import OpenAI
import os
import re
import subprocess
class ShortsGenerator(object):
def read_srt(self,file_path):
subtitles = pysrt.open(file_path)
return subtitles
def extract_text(self,subtitles):
text = ''
for subtitle in subtitles:
text += subtitle.text + ' '
return text.strip()
def get_important_scenes(self,text):
# Load OpenAI API key
client = OpenAI(api_key=os.getenv('OPEN_AI_API_KEY'))
response = client.chat.completions.create(
model="gpt-3.5-turbo",
messages=[
{"role": "system", "content": "You are a helpful videos editing assistant."},
{"role": "user", "content": "Identify the important scenes from the following subtitles text return that by start times and end time,videos should be at less 30s and maximum 2 min with format like this \"1. Arrival of Raymond Reddington at the FBI office - Start time: 00:00:39, End time: 00:01:17\":\n" + text}
],
max_tokens=1500
)
# print(f" this out put : {response.choices[0].message.content}")
important_scenes = response.choices[0].message
return important_scenes
def execute(self,srt_file_path):
subtitles = self.read_srt(srt_file_path)
text = self.extract_text(subtitles)
important_scenes = self.get_important_scenes(text)
return important_scenes
def extract_scenes(self,input_text):
scenes = []
pattern = r'(?P<scene>\d+)\. (?P<description>.*?) - Start time: (?P<start>\d{2}:\d{2}:\d{2}), End time: (?P<end>\d{2}:\d{2}:\d{2})'
matches = re.finditer(pattern, input_text)
for match in matches:
scene_data = match.groupdict()
scenes.append(scene_data)
return scenes
def extract_video_scenes(self,video_file, scenes):
shorts_files_path_list = []
# Output directory
output_dir = "output/"
# Ensure output directory exists
os.makedirs(output_dir, exist_ok=True)
# Process each scene
for scene in scenes:
start_time = scene['start']
end_time = scene['end']
description = scene['description']
output_filename = os.path.join(output_dir, f"{description}.mp4")
shorts_files_path_list.append(output_filename)
# ffmpeg command to extract scene
cmd = [
'ffmpeg',
'-i', video_file,
'-ss', start_time,
'-to', end_time,
'-c:v', 'libx264',
'-c:a', 'aac',
'-strict', 'experimental',
'-b:a', '192k',
output_filename,
'-y' # Overwrite output file if exists
]
subprocess.run(cmd, capture_output=True)
return shorts_files_path_list
if __name__ == "__main__":
srt_file_path = 's1.srt'
path_video = '1.mp4'
shorts_generator = ShortsGenerator()
important_scenes = shorts_generator.execute(srt_file_path)
print("Important Scenes:\n", shorts_generator.extract_scenes(important_scenes.content))
shorts_generator.extract_video_scenes( path_video, shorts_generator.extract_scenes(important_scenes.content))
print("Well Done")
|