Update app.py

app.py

@@ -56,40 +56,19 @@ CACHE_EXAMPLES = torch.device('cuda') and os.getenv("CACHE_EXAMPLES", "0") == "1
 device = torch.device('cuda')
 #device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
-def transcribe(file_upload, progress=gr.Progress(track_tqdm=True)):
+def transcribe(file_upload, progress=gr.Progress(track_tqdm=True)): # microphone
 
-    file = file_upload # microphone if microphone is not None else 
+    file = file_upload # microphone if microphone is not None else
     start_time = time.time()
-    
-    # -- ex subrosa 
-    audio, sr = librosa.load(file_upload, sr=None)
-    duration = librosa.get_duration(y=audio, sr=sr)
-    
-    # -- asr pipeline
-    with torch.no_grad():
-        pipe = pipeline("automatic-speech-recognition", model="NbAiLab/nb-whisper-large", chunk_length_s=30, device=device)
-    
-    # -- process audio in chunks of 30 seconds
-    chunk_size = sr * 30  # 30 seconds
-    text = ""
-    for start in range(0, len(audio), chunk_size):
-        end = min(start + chunk_size, len(audio))
-        chunk = audio[start:end]
-        
-        # -- convert audio chunk to format for pipeline
-        chunk_file = "temp_chunk.wav"
-        sf.write(chunk_file, chunk, sr)
-
 
+    #--------------____________________________________________--------------"
 
-        import librosa
-
+    with torch.no_grad():
+        pipe = pipeline("automatic-speech-recognition", model="NbAiLab/nb-whisper-large", chunk_length_s=30, device=device)
 
+    text = pipe(file)["text"]
 
-    
-        # -- chnk ad transcriptrauma
-        chunk_text = pipe(chunk_file)["text"]
-        text += chunk_text + " "
+    #--------------____________________________________________--------------"
 
     end_time = time.time()
     output_time = end_time - start_time
@@ -101,21 +80,21 @@ def transcribe(file_upload, progress=gr.Progress(track_tqdm=True)):
     memory = psutil.virtual_memory()
     
     # --cpu metric
-    #cpu_usage = psutil.cpu_percent(interval=1)
+    cpu_usage = psutil.cpu_percent(interval=1)
 
     # --gpu metric
-    #gpu_utilization, gpu_memory = GPUInfo.gpu_usage()
+    gpu_utilization, gpu_memory = GPUInfo.gpu_usage()
     
     # --system info string
     system_info = f"""
-    #Processing time: {output_time:.2f} seconds.
-    #Number of words: {word_count}
+    Processing time: {output_time:.2f} seconds.
+    Number of words: {word_count}
+    CPU Usage: {cpu_usage}%
     GPU Memory: {gpu_memory}%
     GPU Utilization: {gpu_utilization}%
     """
-    
-    return text.strip(), system_info
 
+    return text.strip(), system_info
 
 ###############################################################################
 # Interface.
@@ -186,7 +165,7 @@ with iface:
         with gr.Column(scale=3):
             text_output = gr.Textbox(label="Transkribert Tekst", elem_id="transcription_output")
         with gr.Column(scale=1):
-            system_info = gr.Textbox(label="Antall sekunder, ord:", elem_id="system_info_box")
+            system_info = gr.Textbox(label="Antall sekunder, ord, system data:", elem_id="system_info_box")
         
     
     with gr.Tabs():