Update app.py

app.py

@@ -18,7 +18,6 @@ def translate_audio(filepath):
         gr.Warning("No audio found, please retry.")
         return ""
 
-    # Load audio using pydub for chunking
     audio = AudioSegment.from_file(filepath)
     chunk_length_ms = 30 * 1000  # 30 seconds
     chunks = [audio[i:i + chunk_length_ms] for i in range(0, len(audio), chunk_length_ms)]
@@ -29,11 +28,9 @@ def translate_audio(filepath):
         chunk_path = f"chunk_{i}.wav"
         chunk.export(chunk_path, format="wav")
 
-        # Load chunk for model input
         waveform, sample_rate = torchaudio.load(chunk_path)
         inputs = processor(waveform[0], sampling_rate=sample_rate, return_tensors="pt")
 
-        # Generate translated output
         with torch.no_grad():
             generated_ids = model.generate(
                 inputs["input_features"],
@@ -50,27 +47,26 @@ def translate_audio(filepath):
 # Gradio UI components
 mic_transcribe = gr.Interface(
     fn=translate_audio,
-    inputs=gr.Audio(sources="microphone",
-                    type="filepath"),
-    outputs=gr.Textbox(label="Translation (English to Spanish)",
-                       lines=3),
-    allow_flagging="never")
+    inputs=gr.Audio(sources="microphone", type="filepath"),
+    outputs=gr.Textbox(label="Translation (English to Spanish)", lines=3),
+    allow_flagging="never"
+)
 
 file_transcribe = gr.Interface(
     fn=translate_audio,
-    inputs=gr.Audio(sources="upload",
-                    type="filepath"),
-    outputs=gr.Textbox(label="Translation (English to Spanish)",
-                       lines=3),
-    allow_flagging="never",)
+    inputs=gr.Audio(sources="upload", type="filepath"),
+    outputs=gr.Textbox(label="Translation (English to Spanish)", lines=3),
+    allow_flagging="never"
+)
+
+# Initialize Blocks properly
+demo = gr.Blocks()
 
 with demo:
     gr.TabbedInterface(
-        [mic_transcribe,
-         file_transcribe],
-        ["Transcribe Microphone",
-         "Transcribe Audio File"],
+        [mic_transcribe, file_transcribe],
+        ["Transcribe Microphone", "Transcribe Audio File"]
     )
+
 server_port = int(os.environ.get("PORT", 7860))
-demo.launch(share=True, 
-            server_port=server_port)
+demo.launch(share=True, server_port=server_port)