"LOL"

app.py

@@ -21,8 +21,8 @@ genai.configure(api_key=GOOGLE_API_KEY)
 client = InferenceClient(provider="hf-inference", api_key=HF_TOKEN)
 
 # Định nghĩa đường dẫn video đầu ra
-OUTPUT_VIDEO_PATH = "./final_output.mp4"
-OUTPUT_VIDEO_FIXED_PATH = "./final_output_fixed.mp4"
+OUTPUT_VIDEO_PATH = "final_output.mp4"
+OUTPUT_VIDEO_FIXED_PATH = "final_output_fixed.mp4"
 
 # Tiêu đề ứng dụng
 st.set_page_config(page_title="KnowFlow", page_icon="📖")
@@ -41,7 +41,7 @@ uploaded_file = st.file_uploader("📂 Upload your document (PDF)", type=["pdf",
 # Nếu có file, lưu vào thư mục tạm và lấy đường dẫn
 file_path = None
 if uploaded_file:
-    file_path = f"./{uploaded_file.name}"
+    file_path = f"{uploaded_file.name}"
     with open(file_path, "wb") as f:
         f.write(uploaded_file.getbuffer())  # Lưu file thực tế
 number_of_images = st.slider("🖼️ Nhập số ảnh",1,10,3)

main.py

@@ -3,11 +3,8 @@ from src.text_to_speech import text_to_speech
 from src.image_gen import image_gen
 from src.text_to_video import text_to_video
 import os
-def main(file_path = "./data/input/sample.pdf", analysis_level='basic', writting_style='academic', word_lower_limit=100, word_upper_limit = 150, gender = "female", speed = "fast", number_of_images = 3, detail_level="short", perspective="neutral", emotion="sad", time_setting="classic", art_style="realistic", style="anime", color_palette="monochrome"):
+def main(file_path, analysis_level='basic', writting_style='academic', word_lower_limit=100, word_upper_limit = 150, gender = "female", speed = "fast", number_of_images = 3, detail_level="short", perspective="neutral", emotion="sad", time_setting="classic", art_style="realistic", style="anime", color_palette="monochrome"):
     text_processing(file_path = file_path, analysis_level=analysis_level, writting_style=writting_style, word_lower_limit = word_lower_limit, word_upper_limit=word_upper_limit )
     text_to_speech(gender = gender, speed = speed)
     image_gen(number_of_images = number_of_images, detail_level=detail_level, perspective=perspective, emotion=emotion, time_setting=time_setting, art_style=art_style, style=style, color_palette=color_palette)
-    text_to_video()
-
-if __name__ == "__main__":
-    main(file_path="./data/input/sample_2.pdf")
+    text_to_video()

src/image_gen.py

@@ -16,7 +16,7 @@ genai.configure(api_key=GOOGLE_API_KEY)
 client = InferenceClient(provider="hf-inference", api_key=HF_TOKEN)
 
 def split_text_for_images(number_of_images):
-    with open("./text.txt", "r", encoding="utf-8") as file:
+    with open("text.txt", "r", encoding="utf-8") as file:
         text = file.read().strip()
 
     total_length = len(text)
@@ -101,7 +101,7 @@ def image_gen(number_of_images = 3,detail_level = "short", perspective="neutral"
     texts = split_text_for_images(number_of_images)
     index = 0
     for text in tqdm(texts, desc="Processing", unit="image"):
-        output_path = f"./{index}.png"
+        output_path = f"{index}.png"
         prompt = describe_image(text, detail_level, perspective, emotion, time_setting, art_style)
         print(prompt)
 
@@ -120,6 +120,4 @@ def image_gen(number_of_images = 3,detail_level = "short", perspective="neutral"
                 wait_time = 2 ** retry_count + random.uniform(0, 1)  # Exponential backoff
                 print(f"Thử lại sau {wait_time:.2f} giây...")
                 time.sleep(wait_time)
-        index += 1
-if __name__ == "__main__":
-    image_gen(number_of_images = 3, detail_level="short", perspective="neutral", emotion="sad", time_setting="classic", art_style="realistic", style="anime", color_palette="monochrome")
+        index += 1

src/text_processing.py

@@ -15,7 +15,7 @@ def extract_text_from_pdf(pdf_path):
     text = ""
     for page_num in range(doc.page_count):
         page = doc.load_page(page_num)
-        text += page.get_text()
+        text += page.get_text() 
     return text
 
 def extract_text_from_docx(docx_path):
@@ -145,7 +145,4 @@ def text_processing(file_path, analysis_level='basic', writting_style='academic'
                 output_file = os.path.join(output_dir, f"{chunk_idx}_{sentence_idx}.txt")  # Tên file dạng "chunkID_sentenceID.txt"
                 with open(output_file, "w", encoding="utf-8") as f:
                     f.write(sentence.replace("*","") + ".")  # Giữ dấu chấm cuối câu
-                print(f"Đã lưu: {output_file}")
-####################### - MAIN CODE - #######################
-if __name__ == "__main__":
-    text_processing(file_path = "./input/sample_3.pdf")
+                print(f"Đã lưu: {output_file}")

src/text_to_speech.py

@@ -42,11 +42,9 @@ def generate_audio(text, filename="output.mp3", gender="female", speed="normal")
         print("⚠️ Giọng không hợp lệ! Chỉ hỗ trợ 'male' hoặc 'female'.")
 def text_to_speech(gender, speed):
     text_folder = "./"
-    text_files = sorted([f for f in os.listdir(text_folder) if f.endswith('.txt') and f != "text.txt"])
+    text_files = sorted([f for f in os.listdir(text_folder) if f.endswith('.txt') and f != "text.txt" and f != "requirements.txt"])
     for text_file in text_files:
-        with open(f"./text/{text_file}", "r", encoding="utf-8") as file:
+        with open(f"{text_file}", "r", encoding="utf-8") as file:
             content = file.read()
         audio_file = text_file.replace("txt","mp3")
-        generate_audio(content, f"./audio/{audio_file}", gender=gender, speed=speed) 
-if __name__ == "__main__":
-    text_to_speech(gender = "female", speed = "fast")
+        generate_audio(content, f"{audio_file}", gender=gender, speed=speed)