Update deployer/gradio_generator.py

deployer/gradio_generator.py

@@ -1,7 +1,29 @@
-# gradio_generator.py - CPU & Spaces-compatible Gradio interface for RoboSage
+# gradio_generator.py - CPU & Spaces-compatible Gradio interface with voice recording for RoboSage
 
+import os
 import gradio as gr
 from deployer.simulator_interface import VirtualRobot
+from openai import OpenAI
+
+# Initialize OpenAI client for transcription
+OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
+if not OPENAI_API_KEY:
+    raise EnvironmentError("Please set the OPENAI_API_KEY environment variable for audio transcription.
+")
+openai_client = OpenAI(api_key=OPENAI_API_KEY)
+
+
+def transcribe_audio(audio_file: str) -> str:
+    """
+    Transcribe recorded audio using OpenAI Whisper API.
+    Returns the transcribed text.
+    """
+    with open(audio_file, "rb") as f:
+        response = openai_client.audio.transcriptions.create(
+            file=f,
+            model="whisper-1"
+        )
+    return response.text
 
 
 def robot_behavior(user_input: str) -> str:
@@ -23,42 +45,56 @@ def robot_behavior(user_input: str) -> str:
     return bot.perform_action("say I'm sorry, I didn't understand that.")
 
 
+def transcribe_and_respond(audio_file: str) -> str:
+    """
+    Transcribe audio input and run robot_behavior.
+    """
+    text = transcribe_audio(audio_file)
+    return robot_behavior(text)
+
+
 def launch_gradio_app(
     title: str = "RoboSage App",
     description: str = "Your robot, your voice."
 ) -> gr.Blocks:
     """
-    Constructs and returns a Gradio Blocks app with two sections:
-      1. Create your robot app (stubbed)
-      2. Test your robot in simulator
+    Constructs and returns a Gradio Blocks app with voice and text input for simulation.
     """
     with gr.Blocks() as demo:
-        gr.Markdown(f"# 🚀 {title}\n\n{description}")
-
-        # Section: App creation (stubbed)
-        with gr.Accordion("🎨 Create your robot app", open=True):
-            idea = gr.Textbox(label="Robot Idea", placeholder="e.g. A friendly greeting robot.")
-            gen_btn = gr.Button("Generate App", key="gen-app-btn")
-            gen_status = gr.Textbox(label="Generation Status", interactive=False)
-            gen_btn.click(
-                fn=lambda i: "✅ App generation pipeline executed.",
-                inputs=[idea],
-                outputs=[gen_status]
-            )
+        gr.Markdown(f"# 🤖 {title}\n\n{description}")
 
-        # Section: Robot simulation
         with gr.Accordion("🤖 Test your robot", open=True):
-            inp = gr.Textbox(
-                label="Speak or Type Command",
+            # Text input section
+            text_input = gr.Textbox(
+                label="Text Command",
                 placeholder="Type 'hello' or 'say Have a great day!'"
             )
-            out = gr.Textbox(label="Robot Response", lines=4, interactive=False)
-            btn = gr.Button("Send", key="simulate-btn")
-            btn.click(fn=robot_behavior, inputs=[inp], outputs=[out])
+            text_btn = gr.Button("Send Text", key="send-text-btn")
+            text_output = gr.Textbox(label="Robot Response", lines=4, interactive=False)
+            text_btn.click(
+                fn=robot_behavior,
+                inputs=[text_input],
+                outputs=[text_output]
+            )
+
+            gr.Markdown("---")
+            # Voice input section
+            audio_input = gr.Audio(
+                source="microphone",
+                type="filepath",
+                label="Record Command"
+            )
+            audio_btn = gr.Button("Send Audio", key="send-audio-btn")
+            audio_output = gr.Textbox(label="Robot Response (via voice)", lines=4, interactive=False)
+            audio_btn.click(
+                fn=transcribe_and_respond,
+                inputs=[audio_input],
+                outputs=[audio_output]
+            )
 
     return demo
 
 
 if __name__ == "__main__":
     app = launch_gradio_app()
-    app.launch()
+    app.launch()