Update app.py

app.py

@@ -9,15 +9,34 @@ def transcribe(audio):
     generated_text = pipe2(text, max_length=50, num_return_sequences=2)[0]['generated_text']
     return text, generated_text
 
-iface = gr.Interface(
-    fn=transcribe, 
-    inputs=gr.Audio(source="microphone", type="filepath"), 
-    outputs=[
-        gr.Textbox(label='Transcribed Speech'), 
-        gr.Textbox(label='Swedish GPT Generated Speech')
-    ],
-    title="Whisper Small Swedish + Swedish GPT",
-    description="Realtime demo for Swedish speech recognition using a fine-tuned Whisper small model & text generation with Swedish GPT.",
-)
-
-iface.launch()
+with gr.Blocks() as demo:
+    #gr.Interface(
+    #    title="Whisper Small Swedish + Swedish GPT",
+    #    description="Realtime demo for Swedish speech recognition using a fine-tuned Whisper small model & text generation with Swedish GPT.",
+    #)
+    with gr.TabItem("Upload from disk"):
+        upload_file = gr.Audio(source="upload", type="filepath",label="Upload from disk")
+        upload_button = gr.Button("Submit for recognition")
+        upload_outputs = [
+            gr.Textbox(label="Recognized speech from uploaded file"),
+            gr.Textbox(label="Swedish-gpt generated speech from uploaded file")
+        ]
+    with gr.TabItem("Record from microphone"):
+        record_file = gr.Audio(source="microphone", type="filepath",label="Record from microphone")
+        record_button = gr.Button("Submit for recognition")
+        record_outputs = [
+            gr.Textbox(label="Recognized speech from recordings"),
+            gr.Textbox(label="Swedish-gpt generated speech from recordings")
+        ]
+    upload_button.click(
+        fn=transcribe,
+        inputs=upload_file
+        outputs=upload_outputs
+    )
+    record_button.click(
+        fn=transcribe,
+        inputs=record_file
+        outputs=record_outputs
+    )
+        
+demo.launch()