Update app.py

app.py

@@ -1,64 +1,60 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-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
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
-
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    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"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+import requests
+import soundfile as sf
+import tempfile
+import os
+
+# URL of your dedicated processing server.
+# Adjust the port or endpoint path if needed.
+SERVER_URL = "http://204.12.245.139:5000/process_audio"
+
+def process_audio(audio):
+    """
+    Receives audio input from the browser (as a tuple: (sample_rate, numpy_array)),
+    writes it to a temporary WAV file, then sends the file to your dedicated server.
+    Expects a JSON response with a key 'transcription' or 'response' that is returned.
+    """
+    if audio is None:
+        return "No audio provided. Please record something."
+    
+    sample_rate, audio_data = audio
+    # Write the audio data to a temporary WAV file.
+    with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as tmp_file:
+        wav_path = tmp_file.name
+        sf.write(wav_path, audio_data, sample_rate)
+    
+    try:
+        with open(wav_path, "rb") as f:
+            files = {"file": f}
+            # Timeout can be adjusted as needed.
+            response = requests.post(SERVER_URL, files=files, timeout=30)
+        if response.status_code == 200:
+            json_data = response.json()
+            # Try to get a 'transcription' key first, then 'response'
+            result = json_data.get("transcription") or json_data.get("response")
+            if not result:
+                result = "Server processed the audio, but did not return a result."
+        else:
+            result = f"Server error {response.status_code}: {response.text}"
+    except Exception as e:
+        result = f"Exception during processing: {e}"
+    finally:
+        os.remove(wav_path)
+    return result
+
+# Create a Gradio interface that captures audio via the browser microphone.
+iface = gr.Interface(
+    fn=process_audio,
+    inputs=gr.Audio(source="microphone", type="numpy", label="Record Your Voice"),
+    outputs=gr.Textbox(label="Server Response"),
+    title="Live AI Call Agent – Browser Mic Frontend",
+    description=(
+        "Record audio using your browser microphone. The audio will be sent to our dedicated "
+        "server for processing with GPU acceleration. Your server should return a transcription or "
+        "an AI-generated response."
+    )
 )
 
-
 if __name__ == "__main__":
-    demo.launch()
+    # Launch the app so that it listens on all interfaces. Hugging Face Spaces uses these settings.
+    iface.launch(server_name="0.0.0.0", server_port=7860)