Update app.py

app.py

@@ -1,12 +1,13 @@
 import streamlit as st
-from streamlit.components.v1 import html
 import requests
 import os
 import base64
 import tempfile
 import io
-from audiorecorder import audiorecorder
-import streamlit.components.v1 as components
+import numpy as np
+import sounddevice as sd
+import librosa
+from scipy.io.wavfile import write
 from PIL import Image
 
 # Hugging Face API Keys
@@ -34,20 +35,6 @@ avatars = {
     "pig": {"desc": "Warm and cheerful. Creates a positive learning atmosphere.", "voice": "Learning is fun! Let's enjoy this together!"},
 }
 
-# Load the Swiper component
-swiper_code = """
-<script type="module">
-    import React from "react";
-    import ReactDOM from "react-dom";
-    import SwiperComponent from "./Swipercomponent.js";
-    function App() {
-        return <SwiperComponent />;
-    }
-    ReactDOM.render(<App />, document.getElementById("root"));
-</script>
-"""
-st.components.v1.html(swiper_code, height=400)
-
 # Function to process speech-to-text
 def speech_to_text(audio_bytes):
     files = {"file": ("audio.wav", audio_bytes, "audio/wav")}
@@ -67,28 +54,49 @@ def text_to_speech(text):
     payload = {"inputs": text}
     response = requests.post(HF_TEXT_TO_SPEECH_API, headers=HEADERS, json=payload)
     return response.content
+
+# Function to record audio using sounddevice
+def record_audio(duration=5, samplerate=16000):
+    st.write("🎙️ Recording... Speak now!")
+    audio_data = sd.rec(int(samplerate * duration), samplerate=samplerate, channels=1, dtype=np.int16)
+    sd.wait()
+    st.write("✅ Recording finished!")
+
+    # Save as WAV file
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmpfile:
+        write(tmpfile.name, samplerate, audio_data)
+        tmpfile.seek(0)
+        return tmpfile.read()
+
 # Streamlit UI
 st.title("🎙️ AI Speaking Pal")
 st.write("Press the button and chat with your pal!")
 
-# Audio Recording Component
-audio_input = st.audio_input("Let's talk!")
+# Start/Stop Conversation Button
+if "conversation_active" not in st.session_state:
+    st.session_state.conversation_active = False
+
+def toggle_conversation():
+    st.session_state.conversation_active = not st.session_state.conversation_active
+
+st.button("🎤 Start/Stop Conversation", on_click=toggle_conversation)
 
-if audio_input is not None:
-    # Read audio data
-    audio_bytes = audio_input.read()
+# Conversation Text Box
+show_text = st.checkbox("Show conversation text")
+if show_text:
+    conversation_box = st.text_area("Conversation:", height=200)
 
-    # Process speech-to-text
+# Handle microphone input
+if st.session_state.conversation_active:
+    audio_bytes = record_audio(duration=5)  # Record for 5 seconds
     user_text = speech_to_text(audio_bytes)
     st.write(f"**You:** {user_text}")
 
-    # Get AI response
     ai_reply = chatbot_response(user_text)
     st.write(f"**AI:** {ai_reply}")
 
-    # Convert AI response to speech
     speech_audio = text_to_speech(ai_reply)
-    st.audio(speech_audio, format="audio/wav")
+    st.audio(io.BytesIO(speech_audio), format="audio/wav")
 
 # Swiper Carousel Component
 avatar_list = list(avatars.keys())
@@ -105,31 +113,4 @@ if st.button("Hear Sample"):
     audio_bytes = text_to_speech(avatar_info["voice"])
     st.audio(io.BytesIO(audio_bytes), format="audio/wav")
 
-# Start/Stop Conversation Button
-if "conversation_active" not in st.session_state:
-    st.session_state.conversation_active = False
-
-def toggle_conversation():
-    st.session_state.conversation_active = not st.session_state.conversation_active
-
-st.button("🎤 Start/Stop Conversation", on_click=toggle_conversation)
-
-# Conversation Text Box
-show_text = st.checkbox("Show conversation text")
-if show_text:
-    conversation_box = st.text_area("Conversation:", height=200)
-
-# Handle microphone input
-if st.session_state.conversation_active:
-    audio = audiorecorder("Click to Speak", "Stop recording")
-    if len(audio) > 0:
-        with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmpfile:
-            tmpfile.write(audio)
-            user_text = speech_to_text(tmpfile.name)
-            st.write(f"**You:** {user_text}")
-            ai_reply = chatbot_response(user_text)
-            st.write(f"**{selected_avatar.capitalize()}:** {ai_reply}")
-            speech_audio = text_to_speech(ai_reply)
-            st.audio(io.BytesIO(speech_audio), format="audio/wav")
-
 st.write("Have fun learning English with your AI pal!")