Update app.py

app.py

@@ -58,7 +58,7 @@ def transcribe(inputs, previous_transcription):
 def clear():
     return ""
 
-with gr.Blocks() as demo:
+with gr.Blocks() as microphone:
     with gr.Column():
         gr.Markdown(f"# Realtime Whisper Large V3 Turbo: \n Transcribe Audio in Realtime. This Demo uses the Checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers.\n Note: The first token takes about 5 seconds. After that, it works flawlessly.")
         with gr.Row():
@@ -71,4 +71,21 @@ with gr.Blocks() as demo:
         input_audio_microphone.stream(transcribe, [input_audio_microphone, output], [output, latency_textbox], time_limit=45, stream_every=2, concurrency_limit=None)
         clear_button.click(clear, outputs=[output])
 
+with gr.Blocks() as flie:
+    with gr.Column():
+        gr.Markdown(f"# Realtime Whisper Large V3 Turbo: \n Transcribe Audio in Realtime. This Demo uses the Checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers.\n Note: The first token takes about 5 seconds. After that, it works flawlessly.")
+        with gr.Row():
+            input_audio_microphone = gr.Audio(sources="upload", type="numpy")
+            output = gr.Textbox(label="Transcription", value="")
+            latency_textbox = gr.Textbox(label="Latency (seconds)", value="0.0", scale=0)
+        with gr.Row():
+            submit_button = gr.Button("Submit")
+            clear_button = gr.Button("Clear Output")
+
+        submit_button.clcik(transcribe, [input_audio_microphone, output], [output, latency_textbox], concurrency_limit=None)
+        clear_button.click(clear, outputs=[output])
+
+with gr.Blocks() as demo:
+    gr.TabbedInterface([microphone, flie], ["Microphone", "Audio file"])
+
 demo.launch()