Update app.py

app.py

@@ -58,11 +58,12 @@ SIDEBAR_INFO = f"""
 </div>
 """
 
-# ------------transcribe section------------
+#                                                       ------------transcribe section------------
+
 
-@spaces.GPU()
 
 # ============ORIGINAL============[convert m4a audio to wav]
+@spaces.GPU()
 def convert_to_wav(audio_file):
     audio = AudioSegment.from_file(audio_file, format="m4a")
     wav_file = "temp.wav"
@@ -79,19 +80,19 @@ def transcribe_audio(audio_file, batch_size=16):
         audio_file = convert_to_wav(audio_file)
     
     with tempfile.NamedTemporaryFile(suffix=".wav") as temp_audio_file:
-        # Copy the contents of the uploaded audio file to the temporary file
+        # --copy contents of uploaded audio file to temporary file
         temp_audio_file.write(open(audio_file, "rb").read())
         temp_audio_file.flush()
-        # Load the audio file using torchaudio
+        # --use torchaudio to load it
         waveform, sample_rate = torchaudio.load(temp_audio_file.name)
-        # Resample the audio to 16kHz
+        # --resample to 16kHz
         resampler = torchaudio.transforms.Resample(sample_rate, 16000)
         waveform = resampler(waveform)
            
     # --convert to mono
     if waveform.ndim > 1:
         waveform = waveform[0, :]
-    # Convert PyTorch tensor NumPy ndarray
+    # Convert tensor@ndarray
     waveform = waveform.numpy()
 
     start_time = time.time()
@@ -126,35 +127,36 @@ def transcribe_audio(audio_file, batch_size=16):
     return text.strip(), system_info
 
 
-#              ------------summary section------------
+#                                                       ------------summaries section------------
 
-
-#          ------------for app integration later------------
+#                                                   [------------for app integration later------------]
 
 @spaces.GPU()
+# --btw, who is doing this...?
 def clean_text(text):
     text = re.sub(r'https?:\/\/.*[\r\n]*', '', text)
     text = re.sub(r'[^\w\s]', '', text)
     text = re.sub(r'\s+', ' ', text).strip()
     return text
 
-nlp = spacy.blank("nb")  # 'nb' ==> codename = Norwegian Bokmål
+# --SpaCy params
+nlp = spacy.blank("nb")  # ---==> codename ("norsk bokmål")
 nlp.add_pipe('sentencizer')
 spacy_stop_words = spacy.lang.nb.stop_words.STOP_WORDS
 
+# --model (has tokenizer?)
 summarization_model = AutoModel.from_pretrained("NbAiLab/nb-bert-large")
-# pipe = pipeline("fill-mask", model="NbAiLab/nb-bert-large")
+# pipe = pipeline("fill-mask", model="NbAiLab/nb-bert-large") -----hm..
 
+# --process text with SpaCy
 @spaces.GPU()
 def preprocess_text(text):
-    # Process the text with SpaCy
     doc = nlp(text)
-    # SpaCy's stop top wrds direct
     stop_words = spacy_stop_words
-    # Filter out stop words
     words = [token.text for token in doc if token.text.lower() not in stop_words]
     return ' '.join(words)
 
+# --model is called to summarize (need to be placed *after* the three styles and call them)                                                  
 @spaces.GPU()
 def summarize_text(text):
     preprocessed_text = preprocess_text(text)
@@ -172,8 +174,10 @@ def build_similarity_matrix(sentences, stop_words):
                 common_words = set(tokens_a) & set(tokens_b)
                 similarity_matrix.add_edge(i, j, weight=len(common_words))
     return similarity_matrix
+#                                                   [------------model needs to be called for these------------]
+
 
-# PageRank
+# --PageRank
 @spaces.GPU()
 def graph_based_summary(text, num_paragraphs=3):
     doc = nlp(text)
@@ -190,7 +194,7 @@ def graph_based_summary(text, num_paragraphs=3):
     ranked_sentences = sorted(((scores[i], sent) for i, sent in enumerate(sentences)), reverse=True)
     return ' '.join([sent for _, sent in ranked_sentences[:num_paragraphs]])
 
-# LexRank
+# --LexRank
 @spaces.GPU()
 def lex_rank_summary(text, num_paragraphs=3, threshold=0.1):
     doc = nlp(text)
@@ -210,7 +214,7 @@ def lex_rank_summary(text, num_paragraphs=3, threshold=0.1):
     ranked_sentences = sorted(((scores[i], s) for i, s in enumerate(sentences)), reverse=True)
     return ' '.join([ranked_sentences[i][1] for i in range(num_paragraphs)])
 
-# TextRank
+# --TextRank
 @spaces.GPU()
 def text_rank_summary(text, num_paragraphs=3):
     doc = nlp(text)
@@ -228,8 +232,11 @@ def text_rank_summary(text, num_paragraphs=3):
     ranked_sentences = sorted(((scores[i], s) for i, s in enumerate(sentences)), reverse=True)
     return ' '.join([ranked_sentences[i][1] for i in range(num_paragraphs)])
 
-iface = gr.Blocks()
+    
 
+#                                                       ------------interface section------------
+
+iface = gr.Blocks()
 with iface:
     gr.HTML(SIDEBAR_INFO)
     gr.Markdown(HEADER_INFO)