Create app.py

app.py

@@ -0,0 +1,38 @@
+import gradio as gr
+import time
+from openai import OpenAI
+from mistral_common.audio import Audio
+from mistral_common.protocol.instruct.messages import (
+    AudioChunk, TextChunk, UserMessage
+)
+
+client = OpenAI(api_key="EMPTY", base_url="http://localhost:8000/v1")
+
+def transcribe_and_summarize(audio):
+    if not audio:
+        return "Please record or upload an audio file."
+    
+    audio_obj = Audio.from_file(audio, strict=False)
+    chunk = AudioChunk.from_audio(audio_obj)
+    text_chunk = TextChunk(text="Please summarize this audio.")
+    user_msg = UserMessage(content=[chunk, text_chunk]).to_openai()
+
+    response = client.chat.completions.create(
+        model="mistralai/Voxtral-Mini-3B-2507",
+        messages=[user_msg],
+        temperature=0.2,
+        top_p=0.95,
+    )
+
+    return response.choices[0].message.content
+
+
+demo = gr.Interface(
+    fn=transcribe_and_summarize,
+    inputs=gr.Audio(source="microphone", type="filepath", label="Speak now or upload audio"),
+    outputs=gr.Textbox(label="Transcription/Summary"),
+    title="Voxtral Mini (Mistral) via vLLM",
+    description="Transcribe and understand speech using Voxtral Mini 3B with vLLM."
+)
+
+demo.launch(server_name="0.0.0.0", server_port=7860)