Spaces:
Runtime error
Runtime error
Upload confusedautoshortvideogen.py
Browse files- confusedautoshortvideogen.py +772 -0
confusedautoshortvideogen.py
ADDED
@@ -0,0 +1,772 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# -*- coding: utf-8 -*-
|
2 |
+
"""ConfusedAutoShortVideoGen.ipynb
|
3 |
+
|
4 |
+
Automatically generated by Colab.
|
5 |
+
|
6 |
+
Original file is located at
|
7 |
+
https://colab.research.google.com/drive/1qGRLgmJahs6-cNBhO_SIsz_yXKz2OqEW
|
8 |
+
"""
|
9 |
+
|
10 |
+
!pip install gradio
|
11 |
+
!pip install gradio_client
|
12 |
+
!pip install whisperx
|
13 |
+
!pip install pydub
|
14 |
+
|
15 |
+
## Menu
|
16 |
+
##script_writing.py
|
17 |
+
|
18 |
+
mscript_input = "what is depression"
|
19 |
+
mscript_music_input = "What is depression"
|
20 |
+
final_video_output = "final_video_output.mp4"
|
21 |
+
musicownpath = '/content/tmp1mbn3d3s.mp4'
|
22 |
+
|
23 |
+
"""Here’s a consistent and well-structured format where each script name aligns with the key variables:
|
24 |
+
|
25 |
+
**script_writing.py**
|
26 |
+
|
27 |
+
input_text (user_input)
|
28 |
+
script_output (scripttxt_output)
|
29 |
+
script_text (text)
|
30 |
+
audio_gen.py
|
31 |
+
|
32 |
+
script_audio_input (scripttxt_input)
|
33 |
+
audio_file (audio_output)
|
34 |
+
audio_format (mp3)
|
35 |
+
music_gen.py
|
36 |
+
|
37 |
+
script_music_input (scripttxt_input)
|
38 |
+
music_file (music_output)
|
39 |
+
music_format (mp3)
|
40 |
+
time_stamp_code.py
|
41 |
+
|
42 |
+
input_audio (audio_input)
|
43 |
+
timestamp_csv (tscsv_output)
|
44 |
+
csv_format (csv)
|
45 |
+
common_words_remover.py
|
46 |
+
|
47 |
+
raw_csv (csv_input)
|
48 |
+
filtered_csv (commoncsv_output)
|
49 |
+
csv_data (csv)
|
50 |
+
giphy_download.py
|
51 |
+
|
52 |
+
input_keywords (commoncsv_input)
|
53 |
+
downloaded_gif (gif_output)
|
54 |
+
gif_format (gif)
|
55 |
+
gif_crop_concat.py
|
56 |
+
|
57 |
+
raw_gif (gif_input)
|
58 |
+
gif_keywords (commoncsv_input)
|
59 |
+
final_video (concatenate_cropped_output)
|
60 |
+
video_format (mp4)
|
61 |
+
video_finalizer.py
|
62 |
+
|
63 |
+
video_input (concatenate_cropped_input)
|
64 |
+
audio_track (audio_input)
|
65 |
+
timestamps (tscsv_input)
|
66 |
+
background_music (music_input)
|
67 |
+
output_video (final_video_output)
|
68 |
+
output_format (mp4)
|
69 |
+
|
70 |
+
### script_writing.py
|
71 |
+
"""
|
72 |
+
|
73 |
+
import csv
|
74 |
+
import re
|
75 |
+
from datetime import datetime
|
76 |
+
from gradio_client import Client
|
77 |
+
|
78 |
+
# Initialize the client with the correct Hugging Face Space
|
79 |
+
client = Client("Abu1998/Meme_finder")
|
80 |
+
|
81 |
+
# Define the system message and input sentence
|
82 |
+
system_message = """Task: Act as a YouTube Shorts content writer.
|
83 |
+
|
84 |
+
Objective: Create engaging, catchy, and trendy scripts for YouTube Shorts videos that are brief, attention-grabbing, and optimized for viral potential.
|
85 |
+
|
86 |
+
Guidelines:
|
87 |
+
|
88 |
+
Each script should be 15-30 seconds long.
|
89 |
+
Use a hook in the first few seconds to capture viewers' attention.
|
90 |
+
Ensure the content is aligned with trending topics, challenges, or popular culture.
|
91 |
+
Incorporate humor, relatable scenarios, or strong emotions to resonate with the audience.
|
92 |
+
End with a clear call-to-action (CTA) like “Follow for more!” or a cliffhanger.
|
93 |
+
Example Flow:
|
94 |
+
|
95 |
+
User Input: “Write a script about the Monday blues.”
|
96 |
+
AI Output:
|
97 |
+
Script: "POV: It’s Monday morning, and you’re already done with the week. [Clip shows someone groggily hitting the snooze button, dragging themselves out of bed]. But wait… there’s coffee. And suddenly, everything’s okay! ☕✨ [Cut to a quick burst of energy with upbeat music]. If you’re just surviving till the weekend, hit that follow button for more relatable vibes!"
|
98 |
+
"""
|
99 |
+
|
100 |
+
# Define the user input (the sentence for which you want to find the main keyword)
|
101 |
+
user_input = mscript_input
|
102 |
+
|
103 |
+
# Make the API call with the specified parameters
|
104 |
+
result = client.predict(
|
105 |
+
message=user_input,
|
106 |
+
system_message=system_message,
|
107 |
+
max_tokens=512,
|
108 |
+
temperature=0.7,
|
109 |
+
top_p=0.95,
|
110 |
+
api_name="/chat"
|
111 |
+
)
|
112 |
+
|
113 |
+
# Extract the script from the result
|
114 |
+
script = result.strip()
|
115 |
+
|
116 |
+
# Function to split script into words
|
117 |
+
def split_into_words(script_text):
|
118 |
+
words = re.findall(r'\w+', script_text) # Find all words
|
119 |
+
return words
|
120 |
+
|
121 |
+
# Convert the script to a list of words
|
122 |
+
words = split_into_words(script)
|
123 |
+
|
124 |
+
|
125 |
+
# Define the file names with timestamp
|
126 |
+
csv_file = f'updates.csv'
|
127 |
+
txt_file = f'script_output'
|
128 |
+
|
129 |
+
# Save to CSV
|
130 |
+
with open(csv_file, mode='w', newline='', encoding='utf-8') as file:
|
131 |
+
writer = csv.writer(file)
|
132 |
+
writer.writerow(['Content', 'Word']) # Headers
|
133 |
+
for word in words:
|
134 |
+
writer.writerow([user_input, word]) # Write each word as a separate row
|
135 |
+
|
136 |
+
print(f"Script generated, split into words, and saved to {csv_file}.")
|
137 |
+
|
138 |
+
# Save to TXT
|
139 |
+
with open(txt_file, mode='w', encoding='utf-8') as file:
|
140 |
+
file.write(script)
|
141 |
+
|
142 |
+
print(f"Script saved to {txt_file}.")
|
143 |
+
|
144 |
+
"""### audio_gen.py"""
|
145 |
+
|
146 |
+
# Install the gradio_client library
|
147 |
+
|
148 |
+
|
149 |
+
from gradio_client import Client
|
150 |
+
from google.colab import files
|
151 |
+
import shutil
|
152 |
+
|
153 |
+
# Initialize the client with the correct Hugging Face Space
|
154 |
+
client = Client("innoai/Edge-TTS-Text-to-Speech")
|
155 |
+
|
156 |
+
# Upload the script file
|
157 |
+
file_path = "/content/script_output"
|
158 |
+
|
159 |
+
# Read the content from the uploaded script file
|
160 |
+
with open(file_path, 'r', encoding='utf-8') as file:
|
161 |
+
text_input = file.read().strip() # Read and strip any extra whitespace
|
162 |
+
|
163 |
+
# Make the API call with the file content as input
|
164 |
+
result = client.predict(
|
165 |
+
text=text_input,
|
166 |
+
voice="en-US-AvaMultilingualNeural - en-US (Female)", # You can change the voice as needed
|
167 |
+
rate=0, # You can adjust the speech rate if needed
|
168 |
+
pitch=0, # You can adjust the pitch if needed
|
169 |
+
api_name="/predict"
|
170 |
+
)
|
171 |
+
|
172 |
+
# Check the result type and content
|
173 |
+
print(result)
|
174 |
+
|
175 |
+
# Extract the local file path from the result
|
176 |
+
audio_file_path = result[0] # Assuming the audio file path is the first element
|
177 |
+
|
178 |
+
# Define the output file name and path
|
179 |
+
output_file_path = "/content/audio_output.mp3"
|
180 |
+
|
181 |
+
# Copy the file to the desired location
|
182 |
+
shutil.copy(audio_file_path, output_file_path)
|
183 |
+
|
184 |
+
# Provide download link for the generated audio file
|
185 |
+
#files.download(output_file_path)
|
186 |
+
|
187 |
+
"""###Music Gen"""
|
188 |
+
|
189 |
+
#from gradio_client import Client
|
190 |
+
#import requests
|
191 |
+
|
192 |
+
# Initialize the client with the correct Hugging Face Space URL
|
193 |
+
# Make sure the URL is correct and points to a valid Gradio app.
|
194 |
+
#client = Client("https://huggingface.co/spaces/facebook/MusicGen") # Updated URL
|
195 |
+
|
196 |
+
# Define the input for the API
|
197 |
+
#input_text = "youtube shorts background music, beatbox, loop"
|
198 |
+
|
199 |
+
# Query the API (fn_index=2 is used based on the provided information)
|
200 |
+
#result = client.predict(
|
201 |
+
# input_text, # str input in 'parameter_16' Dataset component
|
202 |
+
# fn_index=2
|
203 |
+
#)
|
204 |
+
|
205 |
+
# Check the result type
|
206 |
+
#print(result)
|
207 |
+
|
208 |
+
# Extract the URL from the result
|
209 |
+
# Assuming result is a tuple where the audio URL is the second element
|
210 |
+
#audio_url = result[1]
|
211 |
+
|
212 |
+
# Define the output file path
|
213 |
+
#output_file_path = "/content/tmp1mbn3d3s.mp4"
|
214 |
+
|
215 |
+
# Download the audio content from the URL
|
216 |
+
#response = requests.get(audio_url)
|
217 |
+
#audio_content = response.content
|
218 |
+
|
219 |
+
# Save the audio output to the specified path
|
220 |
+
#with open(output_file_path, 'wb') as f:
|
221 |
+
# f.write(audio_content)
|
222 |
+
|
223 |
+
# Provide download link for the generated audio file
|
224 |
+
#from google.colab import files
|
225 |
+
#files.download(output_file_path)
|
226 |
+
|
227 |
+
"""### Time Stamp"""
|
228 |
+
|
229 |
+
!pip install whisperx
|
230 |
+
|
231 |
+
import whisperx
|
232 |
+
import torch
|
233 |
+
import pandas as pd
|
234 |
+
|
235 |
+
# Initialize the WhisperX model
|
236 |
+
device = "cuda" if torch.cuda.is_available() else "cpu"
|
237 |
+
compute_type = "float32" if device == "cpu" else "float16"
|
238 |
+
model = whisperx.load_model("large-v2", device, compute_type=compute_type)
|
239 |
+
|
240 |
+
def transcribe_and_align(audio_file):
|
241 |
+
# Load audio
|
242 |
+
audio = whisperx.load_audio(audio_file)
|
243 |
+
print("Audio loaded successfully.")
|
244 |
+
|
245 |
+
# Transcribe
|
246 |
+
result = model.transcribe(audio, batch_size=16)
|
247 |
+
print("Transcription result:", result)
|
248 |
+
|
249 |
+
# Align transcription
|
250 |
+
model_a, metadata = whisperx.load_align_model(language_code=result["language"], device=device)
|
251 |
+
result = whisperx.align(result["segments"], model_a, metadata, audio, device, return_char_alignments=True)
|
252 |
+
print("Alignment result:", result)
|
253 |
+
|
254 |
+
# Process segments to get word-level timestamps
|
255 |
+
word_segments = []
|
256 |
+
for segment in result["segments"]:
|
257 |
+
for word_info in segment.get("words", []): # Ensure 'words' is used
|
258 |
+
if "word" in word_info and "start" in word_info and "end" in word_info:
|
259 |
+
word_segments.append({
|
260 |
+
"word": word_info["word"],
|
261 |
+
"start": word_info["start"],
|
262 |
+
"end": word_info["end"],
|
263 |
+
"duration": word_info["end"] - word_info["start"]
|
264 |
+
})
|
265 |
+
|
266 |
+
# Debug: Print word segments to check if they are being populated
|
267 |
+
print("Word segments:", word_segments)
|
268 |
+
|
269 |
+
# Convert the word segments to a DataFrame
|
270 |
+
df = pd.DataFrame(word_segments)
|
271 |
+
|
272 |
+
# Save the result to a CSV file
|
273 |
+
output_file = "/content/transcription_with_word_timestamps.csv" # Ensure correct file path
|
274 |
+
df.to_csv(output_file, index=False)
|
275 |
+
|
276 |
+
return output_file
|
277 |
+
|
278 |
+
# Provide the path to your audio file
|
279 |
+
audio_file_path = "/content/audio_output.mp3"
|
280 |
+
# Transcribe and align the audio file
|
281 |
+
output_file = transcribe_and_align(audio_file_path)
|
282 |
+
|
283 |
+
# Print the path to the output file
|
284 |
+
print(f"Word-level transcription with timestamps saved to: {output_file}")
|
285 |
+
|
286 |
+
"""### common_words_remover"""
|
287 |
+
|
288 |
+
# prompt: write a code to drop these common words from output_file word column , COMMON_WORDS = {"the", "and", "is", "in", "to", "of", "a", "with", "for", "on", "it", "as", "at", "by", "an","this", "that", "which", "or", "be", "are", "was", "were", "has", "have", "had", "why", "such","here", "some", "so", "easy"}
|
289 |
+
|
290 |
+
import pandas as pd
|
291 |
+
def drop_common_words(input_file, output_file, common_words):
|
292 |
+
"""
|
293 |
+
Drops rows containing common words in the 'word' column and saves the result to a new CSV file.
|
294 |
+
|
295 |
+
Args:
|
296 |
+
input_file (str): The path to the input CSV file.
|
297 |
+
output_file (str): The path to the output CSV file.
|
298 |
+
common_words (set): A set of common words to be removed.
|
299 |
+
"""
|
300 |
+
df = pd.read_csv(input_file)
|
301 |
+
df['word'] = df['word'].str.lower() # Convert words to lowercase for comparison
|
302 |
+
df = df[~df['word'].isin(common_words)] # Filter out rows with common words
|
303 |
+
df.to_csv(output_file, index=False)
|
304 |
+
|
305 |
+
# Set of common words to drop
|
306 |
+
COMMON_WORDS = {"the", "and", "is", "in", "to", "of", "a", "with", "for", "on", "it", "as", "at", "by", "an","this", "that", "which", "or", "be", "are", "was", "were", "has", "have", "had", "why", "such","here", "some", "so", "easy"}
|
307 |
+
|
308 |
+
# Input and output file paths
|
309 |
+
input_file = "/content/transcription_with_word_timestamps.csv"
|
310 |
+
output_file = "/content/filtered_transcription.csv"
|
311 |
+
|
312 |
+
# Call the function to drop common words
|
313 |
+
drop_common_words(input_file, output_file, COMMON_WORDS)
|
314 |
+
|
315 |
+
print(f"Rows with common words dropped and saved to {output_file}")
|
316 |
+
|
317 |
+
"""### common_words_remover 2nd step"""
|
318 |
+
|
319 |
+
import pandas as pd
|
320 |
+
from pydub import AudioSegment
|
321 |
+
|
322 |
+
def update_dataframe_with_audio_duration(csv_file, audio_file):
|
323 |
+
# Load the CSV file into a DataFrame
|
324 |
+
df = pd.read_csv(csv_file)
|
325 |
+
|
326 |
+
# Calculate the total duration of the audio
|
327 |
+
audio = AudioSegment.from_file(audio_file)
|
328 |
+
total_duration = audio.duration_seconds
|
329 |
+
|
330 |
+
# Drop existing 'end' and 'duration' columns
|
331 |
+
df = df.drop(columns=['end', 'duration'], errors='ignore')
|
332 |
+
|
333 |
+
# Create a new 'end' column with the next 'start' value
|
334 |
+
df['end'] = df['start'].shift(-1)
|
335 |
+
|
336 |
+
# The first row should start with 0.01
|
337 |
+
df.loc[0, 'start'] = 0.01
|
338 |
+
|
339 |
+
# The last row's 'end' should be the total audio duration
|
340 |
+
df.loc[df.index[-1], 'end'] = total_duration
|
341 |
+
|
342 |
+
# Create a new 'duration' column based on the difference between 'start' and 'end'
|
343 |
+
df['duration'] = df['end'] - df['start']
|
344 |
+
|
345 |
+
# Save the updated DataFrame back to CSV, extracting filename and prepending 'updated_'
|
346 |
+
updated_csv_file = 'updated_' + csv_file.split('/')[-1] # Extract filename and prepend 'updated_'
|
347 |
+
df.to_csv(updated_csv_file, index=False)
|
348 |
+
|
349 |
+
print(f"Updated DataFrame saved to: {updated_csv_file}")
|
350 |
+
return updated_csv_file
|
351 |
+
|
352 |
+
# Example usage
|
353 |
+
csv_file = '/content/filtered_transcription.csv'
|
354 |
+
audio_file = musicownpath
|
355 |
+
update_dataframe_with_audio_duration(csv_file, audio_file)
|
356 |
+
|
357 |
+
"""### **Giphy Gif Download**"""
|
358 |
+
|
359 |
+
# prompt: write a code for "/content/dropped_2024-08-21_18-58-34.csv" to use Word column search in giphy api (API_KEY = "KzPlVn6nz6czmjWpPEy6reL52r1H5gs7") search and download in /content/memes this folder name as the word name
|
360 |
+
|
361 |
+
import requests
|
362 |
+
import csv
|
363 |
+
import os
|
364 |
+
|
365 |
+
# Giphy API details
|
366 |
+
API_KEY = "KzPlVn6nz6czmjWpPEy6reL52r1H5gs7"
|
367 |
+
SEARCH_URL = "https://api.giphy.com/v1/gifs/search"
|
368 |
+
|
369 |
+
# CSV and download directory
|
370 |
+
CSV_FILE = "/content/updated_filtered_transcription.csv"
|
371 |
+
DOWNLOAD_DIR = '/content/memes2'
|
372 |
+
|
373 |
+
# Create download directory if it doesn't exist
|
374 |
+
os.makedirs(DOWNLOAD_DIR, exist_ok=True)
|
375 |
+
|
376 |
+
def download_giphy_gif(search_term, filename):
|
377 |
+
"""Downloads a GIF from Giphy based on the search term."""
|
378 |
+
params = {
|
379 |
+
'api_key': API_KEY,
|
380 |
+
'q': search_term,
|
381 |
+
'limit': 1
|
382 |
+
}
|
383 |
+
response = requests.get(SEARCH_URL, params=params)
|
384 |
+
data = response.json()
|
385 |
+
|
386 |
+
if data['data']:
|
387 |
+
gif_url = data['data'][0]['images']['original']['url']
|
388 |
+
gif_response = requests.get(gif_url)
|
389 |
+
|
390 |
+
with open(os.path.join(DOWNLOAD_DIR, filename), 'wb') as f:
|
391 |
+
f.write(gif_response.content)
|
392 |
+
print(f"Downloaded GIF for '{search_term}' as '{filename}'")
|
393 |
+
else:
|
394 |
+
print(f"No GIF found for '{search_term}'")
|
395 |
+
|
396 |
+
# Process the CSV file
|
397 |
+
with open(CSV_FILE, 'r', encoding='utf-8') as file:
|
398 |
+
reader = csv.DictReader(file)
|
399 |
+
for row in reader:
|
400 |
+
word = row['word']
|
401 |
+
filename = f"{word}.gif"
|
402 |
+
download_giphy_gif(word, filename)
|
403 |
+
|
404 |
+
"""### Updated Download gif
|
405 |
+
|
406 |
+
import requests
|
407 |
+
import csv
|
408 |
+
import os
|
409 |
+
|
410 |
+
# Giphy API details
|
411 |
+
API_KEY = "KzPlVn6nz6czmjWpPEy6reL52r1H5gs7"
|
412 |
+
SEARCH_URL = "https://api.giphy.com/v1/gifs/search"
|
413 |
+
|
414 |
+
# CSV and download directory
|
415 |
+
CSV_FILE = "/content/updated_filtered_transcription.csv"
|
416 |
+
DOWNLOAD_DIR = '/content/memes2'
|
417 |
+
|
418 |
+
# Create download directory if it doesn't exist
|
419 |
+
os.makedirs(DOWNLOAD_DIR, exist_ok=True)
|
420 |
+
|
421 |
+
def download_giphy_gif(search_term, index, offset=0):
|
422 |
+
Downloads a GIF from Giphy based on the search term.
|
423 |
+
params = {
|
424 |
+
'api_key': API_KEY,
|
425 |
+
'q': search_term,
|
426 |
+
'limit': 1,
|
427 |
+
'offset': offset
|
428 |
+
}
|
429 |
+
response = requests.get(SEARCH_URL, params=params)
|
430 |
+
data = response.json()
|
431 |
+
|
432 |
+
if data['data']:
|
433 |
+
gif_url = data['data'][0]['images']['original']['url']
|
434 |
+
gif_response = requests.get(gif_url)
|
435 |
+
|
436 |
+
filename = f"{index}.gif"
|
437 |
+
filepath = os.path.join(DOWNLOAD_DIR, filename)
|
438 |
+
|
439 |
+
with open(filepath, 'wb') as f:
|
440 |
+
f.write(gif_response.content)
|
441 |
+
print(f"Downloaded GIF for '{search_term}' as '{filename}'")
|
442 |
+
return gif_url
|
443 |
+
else:
|
444 |
+
print(f"No GIF found for '{search_term}'")
|
445 |
+
return None
|
446 |
+
|
447 |
+
# Process the CSV file
|
448 |
+
with open(CSV_FILE, 'r', encoding='utf-8') as file:
|
449 |
+
reader = csv.DictReader(file)
|
450 |
+
word_counts = {}
|
451 |
+
for index, row in enumerate(reader):
|
452 |
+
word = row['word']
|
453 |
+
if word in word_counts:
|
454 |
+
word_counts[word] += 1
|
455 |
+
else:
|
456 |
+
word_counts[word] = 1
|
457 |
+
|
458 |
+
gif_url = download_giphy_gif(word, index, offset=word_counts[word] - 1)
|
459 |
+
if gif_url:
|
460 |
+
row['link'] = gif_url
|
461 |
+
|
462 |
+
# Write the updated CSV file
|
463 |
+
with open(CSV_FILE, 'w', newline='', encoding='utf-8') as file:
|
464 |
+
writer = csv.DictWriter(file, fieldnames=['word', 'link'])
|
465 |
+
writer.writeheader()
|
466 |
+
writer.writerows([{'word': word, 'link': gif_url} for word, gif_url in [(row['word'], row['link']) for row in reader]])
|
467 |
+
"""
|
468 |
+
|
469 |
+
# prompt: write a code to save /content/updated_filtered_transcription.csv with index
|
470 |
+
"""
|
471 |
+
import pandas as pd
|
472 |
+
df = pd.read_csv('/content/updated_filtered_transcription.csv')
|
473 |
+
df.to_csv('/content/updated_filtered_transcription.csv', index=True)
|
474 |
+
"""
|
475 |
+
|
476 |
+
"""import requests
|
477 |
+
import csv
|
478 |
+
import os
|
479 |
+
|
480 |
+
# Giphy API details
|
481 |
+
API_KEY = "KzPlVn6nz6czmjWpPEy6reL52r1H5gs7"
|
482 |
+
SEARCH_URL = "https://api.giphy.com/v1/gifs/search"
|
483 |
+
|
484 |
+
# CSV and download directory
|
485 |
+
CSV_FILE = "/content/updated_filtered_transcription.csv"
|
486 |
+
DOWNLOAD_DIR = '/content/memes3'
|
487 |
+
|
488 |
+
# Create download directory if it doesn't exist
|
489 |
+
os.makedirs(DOWNLOAD_DIR, exist_ok=True)
|
490 |
+
|
491 |
+
def download_giphy_gif(search_term, index):
|
492 |
+
Downloads a GIF from Giphy based on the search term.
|
493 |
+
params = {
|
494 |
+
'api_key': API_KEY,
|
495 |
+
'q': search_term,
|
496 |
+
'limit': 1
|
497 |
+
}
|
498 |
+
response = requests.get(SEARCH_URL, params=params)
|
499 |
+
data = response.json()
|
500 |
+
|
501 |
+
if data['data']:
|
502 |
+
gif_url = data['data'][0]['images']['original']['url']
|
503 |
+
gif_response = requests.get(gif_url)
|
504 |
+
|
505 |
+
filename = f"{index}.gif"
|
506 |
+
with open(os.path.join(DOWNLOAD_DIR, filename), 'wb') as f:
|
507 |
+
f.write(gif_response.content)
|
508 |
+
print(f"Downloaded GIF for '{search_term}' as '{filename}'")
|
509 |
+
else:
|
510 |
+
print(f"No GIF found for '{search_term}'")
|
511 |
+
|
512 |
+
# Process the CSV file
|
513 |
+
with open(CSV_FILE, 'r', encoding='utf-8') as file:
|
514 |
+
reader = csv.DictReader(file)
|
515 |
+
for index, row in enumerate(reader):
|
516 |
+
word = row['word']
|
517 |
+
download_giphy_gif(word, index)
|
518 |
+
|
519 |
+
import requests
|
520 |
+
import csv
|
521 |
+
import os
|
522 |
+
|
523 |
+
# Giphy API details
|
524 |
+
API_KEY = "KzPlVn6nz6czmjWpPEy6reL52r1H5gs7"
|
525 |
+
SEARCH_URL = "https://api.giphy.com/v1/gifs/search"
|
526 |
+
|
527 |
+
# CSV and download directory
|
528 |
+
CSV_FILE = "/content/updated_filtered_transcription.csv"
|
529 |
+
DOWNLOAD_DIR = '/content/memes3'
|
530 |
+
|
531 |
+
# Create download directory if it doesn't exist
|
532 |
+
os.makedirs(DOWNLOAD_DIR, exist_ok=True)
|
533 |
+
|
534 |
+
def download_giphy_gif(search_term, index):
|
535 |
+
Downloads a GIF from Giphy based on the search term.
|
536 |
+
params = {
|
537 |
+
'api_key': API_KEY,
|
538 |
+
'q': search_term,
|
539 |
+
'limit': 1
|
540 |
+
}
|
541 |
+
response = requests.get(SEARCH_URL, params=params)
|
542 |
+
data = response.json()
|
543 |
+
|
544 |
+
if data['data']:
|
545 |
+
gif_url = data['data'][0]['images']['original']['url']
|
546 |
+
gif_response = requests.get(gif_url)
|
547 |
+
|
548 |
+
filename = f"{index}.gif"
|
549 |
+
filepath = os.path.join(DOWNLOAD_DIR, filename)
|
550 |
+
|
551 |
+
# Check if the file already exists
|
552 |
+
if not os.path.exists(filepath):
|
553 |
+
with open(filepath, 'wb') as f:
|
554 |
+
f.write(gif_response.content)
|
555 |
+
print(f"Downloaded GIF for '{search_term}' as '{filename}'")
|
556 |
+
else:
|
557 |
+
print(f"GIF for '{search_term}' already exists")
|
558 |
+
|
559 |
+
return gif_url
|
560 |
+
else:
|
561 |
+
print(f"No GIF found for '{search_term}'")
|
562 |
+
return None
|
563 |
+
|
564 |
+
# Process the CSV file
|
565 |
+
with open(CSV_FILE, 'r', encoding='utf-8') as file:
|
566 |
+
reader = csv.DictReader(file)
|
567 |
+
rows = list(reader) # Convert reader to a list to ensure it's fully read
|
568 |
+
|
569 |
+
# Check if any rows were read
|
570 |
+
if rows:
|
571 |
+
# Add a "link" column if it doesn't exist
|
572 |
+
if 'link' not in rows[0].keys():
|
573 |
+
for row in rows:
|
574 |
+
row['link'] = ''
|
575 |
+
|
576 |
+
# Download GIFs and update the "link" column
|
577 |
+
for index, row in enumerate(rows):
|
578 |
+
word = row['word']
|
579 |
+
gif_url = download_giphy_gif(word, index)
|
580 |
+
if gif_url:
|
581 |
+
row['link'] = gif_url
|
582 |
+
|
583 |
+
# Write the updated CSV file
|
584 |
+
with open(CSV_FILE, 'w', newline='', encoding='utf-8') as file:
|
585 |
+
writer = csv.DictWriter(file, fieldnames=rows[0].keys())
|
586 |
+
writer.writeheader()
|
587 |
+
writer.writerows(rows)
|
588 |
+
else:
|
589 |
+
print("The CSV file is empty.")
|
590 |
+
|
591 |
+
import os
|
592 |
+
import pandas as pd
|
593 |
+
|
594 |
+
# Read the CSV file
|
595 |
+
df = pd.read_csv('/content/updated_filtered_transcription.csv')
|
596 |
+
|
597 |
+
# Get the list of words from the DataFrame
|
598 |
+
words = df['word'].tolist()
|
599 |
+
|
600 |
+
# Get the list of GIF files in the directory
|
601 |
+
gif_files = [f for f in os.listdir('/content/memes2') if f.endswith('.gif')]
|
602 |
+
|
603 |
+
# Ensure we have enough words for the GIF files
|
604 |
+
if len(words) < len(gif_files):
|
605 |
+
raise ValueError("Not enough words in the CSV file to rename all GIFs.")
|
606 |
+
|
607 |
+
# Iterate through the GIF files and rename them
|
608 |
+
for i, gif_file in enumerate(gif_files):
|
609 |
+
try:
|
610 |
+
# Generate new filename based on index from the CSV
|
611 |
+
new_filename = f"{i+1}.gif" # Index starts from 1
|
612 |
+
|
613 |
+
# Construct the full paths for the old and new filenames
|
614 |
+
old_path = os.path.join('/content/memes2', gif_file)
|
615 |
+
new_path = os.path.join('/content/memes2', new_filename)
|
616 |
+
|
617 |
+
# Rename the file
|
618 |
+
os.rename(old_path, new_path)
|
619 |
+
print(f"Renamed '{gif_file}' to '{new_filename}'")
|
620 |
+
except Exception as e:
|
621 |
+
print(f"Error processing file '{gif_file}': {e}")
|
622 |
+
|
623 |
+
### Concate Gif (incomplete)
|
624 |
+
"""
|
625 |
+
|
626 |
+
# prompt: drop Unnamed: 0 and save the file
|
627 |
+
|
628 |
+
#import pandas as pd
|
629 |
+
#df = pd.read_csv('/content/updated_filtered_transcription.csv')
|
630 |
+
#df = df.drop(columns=['Unnamed: 0'])
|
631 |
+
#df.to_csv('/content/updated_filtered_transcription.csv', index=False)
|
632 |
+
df.head()
|
633 |
+
|
634 |
+
import moviepy.editor as mpe
|
635 |
+
import os
|
636 |
+
import csv
|
637 |
+
|
638 |
+
# CSV and download directory paths
|
639 |
+
CSV_FILE = '/content/updated_filtered_transcription.csv'
|
640 |
+
DOWNLOAD_DIR = '/content/memes2'
|
641 |
+
OUTPUT_VIDEO = 'updated_concatenated_memes.mp4'
|
642 |
+
|
643 |
+
# Get the GIF order and durations from the CSV file
|
644 |
+
gif_order = []
|
645 |
+
durations = {}
|
646 |
+
with open(CSV_FILE, 'r', encoding='utf-8') as file:
|
647 |
+
reader = csv.DictReader(file)
|
648 |
+
for row in reader:
|
649 |
+
gif_filename = row['word'] + '.gif'
|
650 |
+
duration = float(row['duration']) # Ensure this matches the column name in your CSV
|
651 |
+
gif_order.append(gif_filename)
|
652 |
+
durations[gif_filename] = duration
|
653 |
+
|
654 |
+
# Load, crop, and concatenate GIFs
|
655 |
+
clips = []
|
656 |
+
for gif_filename in gif_order:
|
657 |
+
gif_path = os.path.join(DOWNLOAD_DIR, gif_filename)
|
658 |
+
if os.path.exists(gif_path):
|
659 |
+
clip = mpe.VideoFileClip(gif_path).resize(height=480) # Resize to the same height
|
660 |
+
clip = clip.set_fps(24) # Match the frame rate for consistency
|
661 |
+
|
662 |
+
# Crop each GIF to the specified duration from the new CSV
|
663 |
+
max_duration = durations.get(gif_filename, clip.duration) # Use the duration from the CSV or the full clip duration if not found
|
664 |
+
if clip.duration > max_duration:
|
665 |
+
clip = clip.subclip(0, max_duration) # Keep up to the specified duration
|
666 |
+
|
667 |
+
clips.append(clip)
|
668 |
+
else:
|
669 |
+
print(f"Warning: GIF not found: {gif_filename}")
|
670 |
+
|
671 |
+
# Concatenate and save the video
|
672 |
+
if clips:
|
673 |
+
final_clip = mpe.concatenate_videoclips(clips, method="compose")
|
674 |
+
final_clip.write_videofile(OUTPUT_VIDEO, fps=24) # Set fps to match the GIFs
|
675 |
+
print(f"Concatenated video saved as {OUTPUT_VIDEO}")
|
676 |
+
else:
|
677 |
+
print("No GIFs found to concatenate.")
|
678 |
+
|
679 |
+
"""### concate_audio_gif_music"""
|
680 |
+
|
681 |
+
import moviepy.editor as mpe
|
682 |
+
import os
|
683 |
+
|
684 |
+
# File paths
|
685 |
+
video_file = '/content/updated_concatenated_memes.mp4'
|
686 |
+
music_file = musicownpath
|
687 |
+
audio_file = "/content/audio_output.mp3"
|
688 |
+
output_file = '/content/final_output.mp4'
|
689 |
+
|
690 |
+
# Load the video, music, and audio files
|
691 |
+
video_clip = mpe.VideoFileClip(video_file)
|
692 |
+
music_clip = mpe.VideoFileClip(music_file)
|
693 |
+
audio_clip = mpe.AudioFileClip(audio_file)
|
694 |
+
|
695 |
+
# Duration of the video
|
696 |
+
video_duration = video_clip.duration
|
697 |
+
|
698 |
+
# Ensure the music duration matches the video duration
|
699 |
+
if music_clip.duration < video_duration:
|
700 |
+
# Repeat the music to match the video duration
|
701 |
+
n_repeats = int(video_duration // music_clip.duration) + 1
|
702 |
+
music_clip = mpe.concatenate_videoclips([music_clip] * n_repeats).subclip(0, video_duration)
|
703 |
+
elif music_clip.duration > video_duration:
|
704 |
+
music_clip = music_clip.subclip(0, video_duration)
|
705 |
+
|
706 |
+
# Adjust music volume to 50% and keep audio volume at 100%
|
707 |
+
music_clip = music_clip.volumex(0.3) # Reduce music volume to 50%
|
708 |
+
|
709 |
+
# Ensure the audio duration matches the video duration
|
710 |
+
if audio_clip.duration < video_duration:
|
711 |
+
# Repeat the audio to match the video duration
|
712 |
+
n_repeats = int(video_duration // audio_clip.duration) + 1
|
713 |
+
audio_clip = mpe.concatenate_audioclips([audio_clip] * n_repeats).subclip(0, video_duration)
|
714 |
+
elif audio_clip.duration > video_duration:
|
715 |
+
audio_clip = audio_clip.subclip(0, video_duration)
|
716 |
+
|
717 |
+
# Set the audio of the video clip to the adjusted audio
|
718 |
+
video_clip = video_clip.set_audio(audio_clip)
|
719 |
+
|
720 |
+
# Write the final output video with the adjusted music and audio
|
721 |
+
final_clip = video_clip.set_audio(music_clip.audio)
|
722 |
+
final_clip.write_videofile(output_file, codec='libx264', audio_codec='aac')
|
723 |
+
|
724 |
+
print(f"Final video saved as {output_file}")
|
725 |
+
|
726 |
+
import moviepy.editor as mpe
|
727 |
+
import os
|
728 |
+
|
729 |
+
# File paths
|
730 |
+
video_file = '/content/updated_concatenated_memes.mp4'
|
731 |
+
music_file = musicownpath
|
732 |
+
audio_file = "/content/audio_output.mp3"
|
733 |
+
output_file = '/content/final_output2.mp4'
|
734 |
+
|
735 |
+
# Load the video, music, and audio files
|
736 |
+
video_clip = mpe.VideoFileClip(video_file)
|
737 |
+
music_clip = mpe.VideoFileClip(music_file)
|
738 |
+
audio_clip = mpe.AudioFileClip(audio_file)
|
739 |
+
|
740 |
+
# Duration of the video
|
741 |
+
video_duration = video_clip.duration
|
742 |
+
|
743 |
+
# Ensure the music duration matches the video duration
|
744 |
+
if music_clip.duration < video_duration:
|
745 |
+
# Repeat the music to match the video duration
|
746 |
+
n_repeats = int(video_duration // music_clip.duration) + 1
|
747 |
+
music_clip = mpe.concatenate_videoclips([music_clip] * n_repeats).subclip(0, video_duration)
|
748 |
+
elif music_clip.duration > video_duration:
|
749 |
+
music_clip = music_clip.subclip(0, video_duration)
|
750 |
+
|
751 |
+
# Ensure the audio duration matches the video duration
|
752 |
+
if audio_clip.duration < video_duration:
|
753 |
+
# Repeat the audio to match the video duration
|
754 |
+
n_repeats = int(video_duration // audio_clip.duration) + 1
|
755 |
+
audio_clip = mpe.concatenate_audioclips([audio_clip] * n_repeats).subclip(0, video_duration)
|
756 |
+
elif audio_clip.duration > video_duration:
|
757 |
+
audio_clip = audio_clip.subclip(0, video_duration)
|
758 |
+
|
759 |
+
# Adjust music volume to 50% and keep audio volume at 100%
|
760 |
+
music_clip = music_clip.volumex(0.2) # Reduce music volume to 50%
|
761 |
+
|
762 |
+
# Set the audio of the video clip to the adjusted audio
|
763 |
+
video_clip = video_clip.set_audio(audio_clip)
|
764 |
+
|
765 |
+
# Combine the video with adjusted music
|
766 |
+
final_audio = mpe.CompositeAudioClip([music_clip.audio, audio_clip])
|
767 |
+
final_clip = video_clip.set_audio(final_audio)
|
768 |
+
|
769 |
+
# Write the final output video
|
770 |
+
final_clip.write_videofile(output_file, codec='libx264', audio_codec='aac')
|
771 |
+
|
772 |
+
print(f"Final video saved as {output_file}")
|