Spaces:
Runtime error
Runtime error
KarthickAdopleAI
commited on
Commit
•
b741ac4
1
Parent(s):
2d52a11
Update app.py
Browse files
app.py
CHANGED
@@ -9,12 +9,12 @@ import nltk
|
|
9 |
from gtts import gTTS
|
10 |
from sklearn.feature_extraction.text import TfidfVectorizer
|
11 |
from langchain import HuggingFaceHub, PromptTemplate, LLMChain
|
12 |
-
from huggingsound import SpeechRecognitionModel
|
13 |
import gradio as gr
|
14 |
from pytube import YouTube
|
15 |
import requests
|
16 |
import logging
|
17 |
import os
|
|
|
18 |
nltk.download('punkt')
|
19 |
nltk.download('stopwords')
|
20 |
|
@@ -83,7 +83,6 @@ class VideoAnalytics:
|
|
83 |
def generate_video_summary(self) -> str:
|
84 |
"""
|
85 |
Generate a summary of the transcribed video.
|
86 |
-
|
87 |
Returns:
|
88 |
str: Generated summary.
|
89 |
"""
|
@@ -116,7 +115,6 @@ class VideoAnalytics:
|
|
116 |
def generate_topics(self) -> str:
|
117 |
"""
|
118 |
Generate topics from the transcribed video.
|
119 |
-
|
120 |
Returns:
|
121 |
str: Generated topics.
|
122 |
"""
|
@@ -147,7 +145,6 @@ class VideoAnalytics:
|
|
147 |
def translation(self) -> str:
|
148 |
"""
|
149 |
translation from the transcribed video.
|
150 |
-
|
151 |
Returns:
|
152 |
str: translation.
|
153 |
"""
|
@@ -176,11 +173,9 @@ class VideoAnalytics:
|
|
176 |
def format_prompt(self, question: str, data: str) -> str:
|
177 |
"""
|
178 |
Formats the prompt for the language model.
|
179 |
-
|
180 |
Args:
|
181 |
question (str): The user's question.
|
182 |
data (str): The data to be analyzed.
|
183 |
-
|
184 |
Returns:
|
185 |
str: Formatted prompt.
|
186 |
"""
|
@@ -197,7 +192,6 @@ class VideoAnalytics:
|
|
197 |
repetition_penalty=1.0) -> str:
|
198 |
"""
|
199 |
Generates text based on the prompt and transcribed text.
|
200 |
-
|
201 |
Args:
|
202 |
prompt (str): The prompt for generating text.
|
203 |
transcribed_text (str): The transcribed text for analysis.
|
@@ -205,7 +199,6 @@ class VideoAnalytics:
|
|
205 |
max_new_tokens (int): Maximum number of tokens to generate. Default is 5000.
|
206 |
top_p (float): Nucleus sampling parameter. Default is 0.95.
|
207 |
repetition_penalty (float): Penalty for repeating the same token. Default is 1.0.
|
208 |
-
|
209 |
Returns:
|
210 |
str: Generated text.
|
211 |
"""
|
@@ -241,11 +234,9 @@ class VideoAnalytics:
|
|
241 |
def video_qa(self, question: str, model: str) -> str:
|
242 |
"""
|
243 |
Performs video question answering.
|
244 |
-
|
245 |
Args:
|
246 |
question (str): The question asked by the user.
|
247 |
model (str): The language model to be used ("OpenAI" or "Mixtral").
|
248 |
-
|
249 |
Returns:
|
250 |
str: Answer to the user's question.
|
251 |
"""
|
@@ -273,7 +264,6 @@ class VideoAnalytics:
|
|
273 |
def extract_video_important_sentence(self) -> str:
|
274 |
"""
|
275 |
Extract important sentences from the transcribed video.
|
276 |
-
|
277 |
Returns:
|
278 |
str: Extracted important sentences.
|
279 |
"""
|
@@ -315,7 +305,6 @@ class VideoAnalytics:
|
|
315 |
def write_text_files(self, text: str, filename: str) -> None:
|
316 |
"""
|
317 |
Write text to a file.
|
318 |
-
|
319 |
Args:
|
320 |
text (str): Text to be written to the file.
|
321 |
filename (str): Name of the file.
|
@@ -331,10 +320,8 @@ class VideoAnalytics:
|
|
331 |
def Download(self, link: str) -> str:
|
332 |
"""
|
333 |
Download a video from YouTube.
|
334 |
-
|
335 |
Args:
|
336 |
link (str): YouTube video link.
|
337 |
-
|
338 |
Returns:
|
339 |
str: Path to the downloaded video file.
|
340 |
"""
|
@@ -366,11 +353,9 @@ class VideoAnalytics:
|
|
366 |
def main(self, video: str = None, input_path: str = None) -> tuple:
|
367 |
"""
|
368 |
Perform video analytics.
|
369 |
-
|
370 |
Args:
|
371 |
video (str): Path to the video file.
|
372 |
input_path (str): Input path for the video.
|
373 |
-
|
374 |
Returns:
|
375 |
tuple: Summary, important sentences, and topics.
|
376 |
"""
|
@@ -381,19 +366,19 @@ class VideoAnalytics:
|
|
381 |
video_ = VideoFileClip(input_path)
|
382 |
duration = video_.duration
|
383 |
video_.close()
|
384 |
-
if round(duration) <=
|
385 |
text = self.transcribe_video(input_path)
|
386 |
else:
|
387 |
-
return "Video Duration Above 10 Minutes,Try Below 10 Minutes Video","",""
|
388 |
elif video:
|
389 |
video_ = VideoFileClip(video)
|
390 |
duration = video_.duration
|
391 |
video_.close()
|
392 |
-
if round(duration) <=
|
393 |
text = self.transcribe_video(video)
|
394 |
input_path = video
|
395 |
else:
|
396 |
-
return "Video Duration Above 10 Minutes,Try Below 10 Minutes Video","",""
|
397 |
# Generate summary, important sentences, and topics
|
398 |
summary = self.generate_video_summary()
|
399 |
self.write_text_files(summary,"Summary")
|
@@ -454,7 +439,7 @@ class VideoAnalytics:
|
|
454 |
submit_btn.click(self.main,[video,yt_link],[summary,Important_Sentences,Topics,summary_audio,important_sentence_audio,topics_audio])
|
455 |
question.submit(self.video_qa,[question,model],result)
|
456 |
demo.launch()
|
457 |
-
|
458 |
if __name__ == "__main__":
|
459 |
video_analytics = VideoAnalytics()
|
460 |
video_analytics.gradio_interface()
|
|
|
9 |
from gtts import gTTS
|
10 |
from sklearn.feature_extraction.text import TfidfVectorizer
|
11 |
from langchain import HuggingFaceHub, PromptTemplate, LLMChain
|
|
|
12 |
import gradio as gr
|
13 |
from pytube import YouTube
|
14 |
import requests
|
15 |
import logging
|
16 |
import os
|
17 |
+
from huggingsound import SpeechRecognitionModel
|
18 |
nltk.download('punkt')
|
19 |
nltk.download('stopwords')
|
20 |
|
|
|
83 |
def generate_video_summary(self) -> str:
|
84 |
"""
|
85 |
Generate a summary of the transcribed video.
|
|
|
86 |
Returns:
|
87 |
str: Generated summary.
|
88 |
"""
|
|
|
115 |
def generate_topics(self) -> str:
|
116 |
"""
|
117 |
Generate topics from the transcribed video.
|
|
|
118 |
Returns:
|
119 |
str: Generated topics.
|
120 |
"""
|
|
|
145 |
def translation(self) -> str:
|
146 |
"""
|
147 |
translation from the transcribed video.
|
|
|
148 |
Returns:
|
149 |
str: translation.
|
150 |
"""
|
|
|
173 |
def format_prompt(self, question: str, data: str) -> str:
|
174 |
"""
|
175 |
Formats the prompt for the language model.
|
|
|
176 |
Args:
|
177 |
question (str): The user's question.
|
178 |
data (str): The data to be analyzed.
|
|
|
179 |
Returns:
|
180 |
str: Formatted prompt.
|
181 |
"""
|
|
|
192 |
repetition_penalty=1.0) -> str:
|
193 |
"""
|
194 |
Generates text based on the prompt and transcribed text.
|
|
|
195 |
Args:
|
196 |
prompt (str): The prompt for generating text.
|
197 |
transcribed_text (str): The transcribed text for analysis.
|
|
|
199 |
max_new_tokens (int): Maximum number of tokens to generate. Default is 5000.
|
200 |
top_p (float): Nucleus sampling parameter. Default is 0.95.
|
201 |
repetition_penalty (float): Penalty for repeating the same token. Default is 1.0.
|
|
|
202 |
Returns:
|
203 |
str: Generated text.
|
204 |
"""
|
|
|
234 |
def video_qa(self, question: str, model: str) -> str:
|
235 |
"""
|
236 |
Performs video question answering.
|
|
|
237 |
Args:
|
238 |
question (str): The question asked by the user.
|
239 |
model (str): The language model to be used ("OpenAI" or "Mixtral").
|
|
|
240 |
Returns:
|
241 |
str: Answer to the user's question.
|
242 |
"""
|
|
|
264 |
def extract_video_important_sentence(self) -> str:
|
265 |
"""
|
266 |
Extract important sentences from the transcribed video.
|
|
|
267 |
Returns:
|
268 |
str: Extracted important sentences.
|
269 |
"""
|
|
|
305 |
def write_text_files(self, text: str, filename: str) -> None:
|
306 |
"""
|
307 |
Write text to a file.
|
|
|
308 |
Args:
|
309 |
text (str): Text to be written to the file.
|
310 |
filename (str): Name of the file.
|
|
|
320 |
def Download(self, link: str) -> str:
|
321 |
"""
|
322 |
Download a video from YouTube.
|
|
|
323 |
Args:
|
324 |
link (str): YouTube video link.
|
|
|
325 |
Returns:
|
326 |
str: Path to the downloaded video file.
|
327 |
"""
|
|
|
353 |
def main(self, video: str = None, input_path: str = None) -> tuple:
|
354 |
"""
|
355 |
Perform video analytics.
|
|
|
356 |
Args:
|
357 |
video (str): Path to the video file.
|
358 |
input_path (str): Input path for the video.
|
|
|
359 |
Returns:
|
360 |
tuple: Summary, important sentences, and topics.
|
361 |
"""
|
|
|
366 |
video_ = VideoFileClip(input_path)
|
367 |
duration = video_.duration
|
368 |
video_.close()
|
369 |
+
if round(duration) <= 6*600:
|
370 |
text = self.transcribe_video(input_path)
|
371 |
else:
|
372 |
+
return "Video Duration Above 10 Minutes,Try Below 10 Minutes Video","","",None,None,None
|
373 |
elif video:
|
374 |
video_ = VideoFileClip(video)
|
375 |
duration = video_.duration
|
376 |
video_.close()
|
377 |
+
if round(duration) <= 6*600:
|
378 |
text = self.transcribe_video(video)
|
379 |
input_path = video
|
380 |
else:
|
381 |
+
return "Video Duration Above 10 Minutes,Try Below 10 Minutes Video","","",None,None,None
|
382 |
# Generate summary, important sentences, and topics
|
383 |
summary = self.generate_video_summary()
|
384 |
self.write_text_files(summary,"Summary")
|
|
|
439 |
submit_btn.click(self.main,[video,yt_link],[summary,Important_Sentences,Topics,summary_audio,important_sentence_audio,topics_audio])
|
440 |
question.submit(self.video_qa,[question,model],result)
|
441 |
demo.launch()
|
442 |
+
|
443 |
if __name__ == "__main__":
|
444 |
video_analytics = VideoAnalytics()
|
445 |
video_analytics.gradio_interface()
|