Update app.py

app.py

@@ -3,6 +3,7 @@ from kokoro import KPipeline
 import soundfile as sf
 import io
 import os
+from googletrans import Translator  # We will use googletrans for translation
 
 # Install espeak-ng if not installed
 # if not os.system("which espeak-ng"):
@@ -15,7 +16,7 @@ import os
 st.title("Text-to-Speech with Kokoro")
 # Expander section to display information in multiple languages
 with st.expander("Sample Prompt!"):
-    st.markdown("""
+    st.markdown(""" 
     - My name is Shukdev. (In English)
     - Mi nombre es Shukdev. (In Spanish)
     - Je m'appelle Choukdev. (In French)
@@ -57,10 +58,6 @@ st.sidebar.markdown("""
 Enjoy experimenting with the text-to-speech conversion, and feel free to try different voices, speeds, and languages!
 """)
 
-st.sidebar.markdown("""
-        ### Courtesy: [Kokoro](https://huggingface.co/hexgrad/Kokoro-82M?fbclid=IwY2xjawIKqzxleHRuA2FlbQIxMAABHaf9GldgYOzXktNuoRtNKqd-aL7r-S7zPGyC8ttYOiG2zYfQqLyV4Qm75A_aem_0wKLC2C87ZZ2F04WjPJbtA)
-    """)
-
 # User input for text, language, and voice settings
 input_text = st.text_area("Enter your text here", placeholder="The sky above the port was the color of television...")
 lang_code = st.selectbox("Select Language", ['a', 'b', 'e', 'f', 'h', 'i', 'p', 'z', 'j'])
@@ -87,6 +84,9 @@ speed = st.slider("Speed", min_value=0.5, max_value=2.0, value=1.0, step=0.1)
 # Initialize the TTS pipeline with user-selected language
 pipeline = KPipeline(lang_code=lang_code)
 
+# Initialize translator
+translator = Translator()
+
 # Generate Audio function
 def generate_audio(text, lang_code, voice, speed):
     generator = pipeline(text, voice=voice, speed=speed, split_pattern=r'\n+')
@@ -99,6 +99,13 @@ def generate_audio(text, lang_code, voice, speed):
         buffer.seek(0)
         return buffer
 
+# Translate text to English
+def translate_to_english(text, lang_code):
+    if lang_code != 'a':  # Only translate if it's not already English
+        translation = translator.translate(text, src=lang_code, dest='en')
+        return translation.text
+    return text
+
 # Generate and display the audio file
 if st.button('Generate Audio'):
     st.write("Generating speech...")
@@ -109,8 +116,26 @@ if st.button('Generate Audio'):
 
     # Optional: Save the generated audio file for download
     st.download_button(
-        label="Download Audio",
+        label="Download Audio (Original Text)",
         data=audio_buffer,
-        file_name="generated_speech.wav",
+        file_name="generated_speech_original.wav",
+        mime="audio/wav"
+    )
+
+    # Translate the input text to English
+    translated_text = translate_to_english(input_text, lang_code)
+
+    # Generate audio for the translated English text
+    translated_audio_buffer = generate_audio(translated_text, 'a', voice, speed)
+
+    # Display Audio for the translated text
+    st.write(f"Translated Text: {translated_text}")
+    st.audio(translated_audio_buffer, format='audio/wav')
+
+    # Download option for the translated audio
+    st.download_button(
+        label="Download Audio (Translated to English)",
+        data=translated_audio_buffer,
+        file_name="generated_speech_translated.wav",
         mime="audio/wav"
-    )
+    )