Update app.py

app.py

@@ -3,7 +3,7 @@ import gradio as gr
 from numpy import random 
 from PIL import Image
 pipe = pipeline(model="fimster/whisper-small-sv-SE")  # change to "your-username/the-name-you-picked"
-images = ["katt", "melon", "hund", "banan"]
+images = ["katt", "melon", "banan"]
 image = random.choice(images)
 
 def transcribe(audio, img):
@@ -23,7 +23,7 @@ iface = gr.Interface(
     inputs=[gr.Audio(source="microphone", type="filepath"), gr.Image("./images/" + image + ".jpeg")], 
     outputs="text",
     title="Whisper Small Swedish",
-    description="Proof of concept for Duolingo-type gamified language learning",
+    description="Proof of concept for Duolingo-type gamified language learning. Record and say what the thing in the picture is, then click submit and the model will check if you were correct!",
 )
 
 iface.launch()