Create app.py

app.py

@@ -0,0 +1,66 @@
+import streamlit as st
+from utils import save_uploaded_audio
+from voice_cloner.py import clone_and_generate_text
+import os
+import whisper
+import torchaudio
+from groq import Groq
+
+# Load Whisper model
+whisper_model = whisper.load_model("tiny")
+
+# Groq API client
+groq_client = Groq(api_key=os.environ.get("GROQ_API_KEY"))
+
+st.set_page_config(page_title="Voice Cloner Chat", layout="centered")
+st.title("🎙️ Voice Cloner Chat App")
+
+# Session State
+if "clone_path" not in st.session_state:
+    st.session_state.clone_path = None
+
+st.sidebar.header("Voice Input")
+option = st.sidebar.radio("Choose input method", ["Upload Voice", "Record Voice"])
+
+if option == "Upload Voice":
+    uploaded = st.sidebar.file_uploader("Upload voice sample", type=["wav", "mp3"])
+    if uploaded:
+        voice_path = save_uploaded_audio(uploaded, "reference_voice.wav")
+        st.session_state.clone_path = voice_path
+        st.success("Voice uploaded and saved as clone voice.")
+
+if option == "Record Voice":
+    duration = st.sidebar.slider("Duration (seconds)", 3, 10, 5)
+    if st.sidebar.button("Record"):
+        from utils import record_audio
+        path = record_audio(duration=duration)
+        st.session_state.clone_path = path
+        st.success("Voice recorded and saved as clone voice.")
+
+st.divider()
+st.subheader("💬 Talk with AI using cloned voice")
+
+user_voice = st.file_uploader("Upload your voice question", type=["wav", "mp3"])
+if user_voice:
+    user_voice_path = save_uploaded_audio(user_voice, "user_question.wav")
+    st.audio(user_voice_path)
+
+    # Transcribe question
+    result = whisper_model.transcribe(user_voice_path)
+    user_text = result["text"]
+    st.write(f"📝 Transcribed Text: {user_text}")
+
+    # Generate response using Groq
+    response = groq_client.chat.completions.create(
+        model="llama3-8b-8192",
+        messages=[{"role": "user", "content": user_text}]
+    )
+    reply = response.choices[0].message.content
+    st.write(f"🤖 AI Response: {reply}")
+
+    if st.checkbox("Use cloned voice to reply"):
+        if st.session_state.clone_path:
+            clone_audio = clone_and_generate_text(reply, st.session_state.clone_path)
+            st.audio(clone_audio)
+        else:
+            st.warning("No clone voice found. Please upload or record one.")