Update app.py

app.py

@@ -1,14 +1,13 @@
-import gradio as gr
+import streamlit as st
 from transformers import pipeline
-import pyttsx3
-import speech_recognition as sr
+import soundfile as sf
 
 # Initialize translation model (English <-> Urdu)
 translation_model = pipeline("translation_en_to_ur", model="Helsinki-NLP/opus-mt-en-ur")
 reverse_translation_model = pipeline("translation_ur_to_en", model="Helsinki-NLP/opus-mt-ur-en")
 
-# Initialize text-to-speech engine
-engine = pyttsx3.init()
+# Initialize text-to-speech model from Hugging Face
+tts_model = pipeline("text-to-speech", model="facebook/fastspeech2-en-ljspeech")
 
 # Function to translate text and provide feedback
 def translate_and_speak(text, direction):
@@ -17,39 +16,27 @@ def translate_and_speak(text, direction):
     else:
         translated_text = reverse_translation_model(text)[0]['translation_text']
     
-    # Use TTS to speak the translated text
-    engine.say(translated_text)
-    engine.runAndWait()
-    
-    return translated_text
+    # Use TTS to synthesize speech from translated text
+    audio = tts_model(translated_text)
+    audio_path = "output.wav"
+    sf.write(audio_path, audio["array"], 22050)  # Save audio to file
+    return translated_text, audio_path
 
-# Function for Speech-to-Text
-def speech_to_text(audio_file, direction):
-    # Recognize speech from audio
-    recognizer = sr.Recognizer()
-    audio = sr.AudioFile(audio_file.name)
-    with audio as source:
-        audio_data = recognizer.record(source)
-    
-    try:
-        text = recognizer.recognize_google(audio_data)
-    except sr.UnknownValueError:
-        text = "Sorry, could not understand the audio."
-    except sr.RequestError:
-        text = "Could not request results from Google Speech Recognition service."
-    
-    # Translate text
-    return translate_and_speak(text, direction)
+# Streamlit app UI
+st.title("AI-Powered Language Tutor")
+st.write("An interactive tutor to help you practice English-Urdu translations with speech feedback!")
 
-# Gradio interface
-iface = gr.Interface(
-    fn=translate_and_speak, 
-    inputs=[gr.Textbox(label="Enter Text"), 
-            gr.Radio(choices=["English to Urdu", "Urdu to English"], label="Translation Direction")],
-    outputs="text", 
-    live=True,
-    title="AI-Powered Language Tutor",
-    description="An interactive tutor to help you practice English-Urdu translations with speech feedback!"
-)
+# User input for translation
+text_input = st.text_area("Enter Text", "Hello, how are you?")
+direction = st.radio("Choose Translation Direction", ["English to Urdu", "Urdu to English"])
 
-iface.launch()
+# Button to process the text and play audio
+if st.button("Translate and Speak"):
+    translated_text, audio_path = translate_and_speak(text_input, direction)
+    
+    # Display translated text and audio
+    st.subheader("Translated Text:")
+    st.write(translated_text)
+    
+    st.subheader("Generated Speech:")
+    st.audio(audio_path, format="audio/wav")