Update app.py

app.py

@@ -1,13 +1,12 @@
+# https://huggingface.co/spaces/FlavioBF/AI_in_production_PRJs
+
 # ================================================================
-# TESTING VERSION
-# ALL-IN-ONE CELL VERSION
-# OF THE PROGRAM
+# 
+# import
+# 
 # ================================================================
-#
-# -------------------------
-# PDF
-# -------------------------
 
+#PDF PROCESSING
 # To read the PDF
 import PyPDF2
 # To analyze the PDF layout and extract text
@@ -20,9 +19,19 @@ from PIL import Image
 from pdf2image import convert_from_path
 # To perform OCR to extract text from images
 import pytesseract
+
 # To remove the additional created files
 import os
 
+#SUMMARIZATION AND AUDIO PROCESSING
+import torch
+import numpy as np
+import scipy
+import gradio as gr
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+from transformers import pipeline, AutoProcessor, AutoModel
+from transformers import pipeline 
+
 # -----------------------------------------------------------------------------
 # Create a function to extract text
 
@@ -263,12 +272,7 @@ pdf_path2=os.path.join(os.path.abspath(""), "1812_05944.pdf")
 # 
 # ======================================= 
 def sentence_to_audio(fileobj):
-    from transformers import pipeline, AutoProcessor, AutoModel
-    import torch
-    from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
-    from transformers import pipeline 
-    import numpy as np
-    import scipy
+
 
     
     # text mining from pdf
@@ -308,38 +312,19 @@ def sentence_to_audio(fileobj):
     scipy.io.wavfile.write("s_2_s.wav", rate=generated_audio["sampling_rate"], data=generated_audio["audio"].T)
     return "s_2_s.wav",summary_text    
 
-
-
-    
-#    processor = AutoProcessor.from_pretrained("suno/bark-small")
-#    model = AutoModel.from_pretrained("suno/bark-small")
-#    inputs = processor(
-#        text=summary_text,
-#        return_tensors="pt",
-#    )
-#    speech_values = model.generate(**inputs, do_sample=True)
-#    sampling_rate = model.generation_config.sample_rate
-#    return sampling_rate, speech_values.cpu().numpy().squeeze(),summary_text
-
-
-# ============================================================================================
-
-
-# =======================================
-
-import gradio as gr
-from transformers import pipeline, AutoProcessor, AutoModel
-from transformers import pipeline
-
 # ===========================================================    
 
 #summary_txt="It is dangerous to think of machine learning as a free-to-use toolkit, as it is common to incur ongoing maintenance costs in real-world ML systems"
-#sentence_to_audio(summary_txt)
 
 pdf_path=os.path.join(os.path.abspath(""), "hidden-technical-debt-in-machine-learning-systems-Paper.pdf")
 #pdf_path2=os.path.join(os.path.abspath(""), "1812_05944.pdf")
 pdf_path2=os.path.join(os.path.abspath(""), "Article_4_ExperimentalEvidence_on_the_Productivity_Effects_ of_Generative_ Artificial_Intelligence.pdf")
 
 
-demo = gr.Interface(fn=sentence_to_audio, inputs="file", outputs=["audio","text"],examples=[pdf_path,pdf_path2])
+
+#iface = gr.Interface(fn=sentence_to_audio, inputs="file", outputs=["audio",gr.Textbox(lines=4,label="one sentence summ.")],title="SINGLE SENTENCE SUMMARY TO AUDIO CONVERSIONE (upload only pdf files with Abstract section)")
+#iface.launch(share=True)
+
+
+demo = gr.Interface(fn=sentence_to_audio, inputs="file", outputs=["audio",,gr.Textbox(lines=4,label="one sentence summ.")],examples=[pdf_path,pdf_path2],title="SINGLE SENTENCE SUMMARY TO AUDIO CONVERSIONE (upload only pdf files with Abstract section)")
 demo.launch(share=True)