whisper

app.py

@@ -1,5 +1,6 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
+import whisper
 
 """
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
@@ -15,9 +16,16 @@ def respond(
     temperature,
     top_p,
 ):
-    # Here you could process the audio file if needed
-    # For now, just returning our fixed response
-    yield "Buang NB"
+    # Load whisper model (you might want to move this outside the function)
+    model = whisper.load_model("base")
+    
+    # Transcribe the audio file
+    if audio is not None:
+        result = model.transcribe(audio)
+        transcribed_text = result["text"]
+        yield f"Transcribed: {transcribed_text}\nResponse: Buang NB"
+    else:
+        yield "No audio detected. Please try again."
 
 
 """
@@ -26,7 +34,7 @@ For information on how to customize the ChatInterface, peruse the gradio docs: h
 demo = gr.ChatInterface(
     respond,
     chatbot=gr.Chatbot(),
-    textbox=gr.Audio(source="microphone", type="filepath"),  # Changed to Audio input
+    textbox=gr.Audio(type="filepath"),  # Removed 'source' parameter as it's not supported
     additional_inputs=[
         gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),