Update app.py

app.py

@@ -1,12 +1,10 @@
 import streamlit as st
-from streamlit_webrtc import webrtc_streamer, WebRtcMode
-import av
+from streamlit_mic import st_mic
 import whisper
 import openai
+from gtts import gTTS
 import tempfile
 import os
-from gtts import gTTS
-from pydub import AudioSegment
 
 # Set your API Key (Groq-compatible)
 openai.api_key = os.getenv("GROQ_API_KEY", "your-groq-api-key")
@@ -18,30 +16,17 @@ st.title("🎙️ Voice-to-Voice Conversational App")
 
 st.info("🎤 Record your voice and click 'Stop' to process:")
 
-webrtc_ctx = webrtc_streamer(
-    key="example",
-    mode=WebRtcMode.SENDRECV,
-    audio_receiver_size=1024,
-    media_stream_constraints={"audio": True, "video": False},
-)
-
-if "audio_bytes" not in st.session_state:
-    st.session_state.audio_bytes = b""
-
-if webrtc_ctx.audio_receiver:
-    audio_frames = webrtc_ctx.audio_receiver.get_frames(timeout=1)
-    for frame in audio_frames:
-        st.session_state.audio_bytes += frame.to_ndarray().tobytes()
+audio_data = st_mic()
 
-if st.button("🛑 Process Voice"):
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as f:
-        f.write(st.session_state.audio_bytes)
-        audio_path = f.name
+if audio_data:
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp:
+        tmp.write(audio_data)
+        tmp_path = tmp.name
 
-    st.audio(audio_path)
+    st.audio(tmp_path)
 
     st.info("🔠 Transcribing...")
-    result = model.transcribe(audio_path)
+    result = model.transcribe(tmp_path)
     user_input = result["text"]
     st.success(f"You said: {user_input}")