Update app.py

app.py

@@ -3,8 +3,6 @@ from kokoro import KPipeline
 import soundfile as sf
 import io
 import os
-import speech_recognition as sr
-import romkan  # For Japanese Romanization
 
 # Install espeak-ng if not installed
 if not os.system("which espeak-ng"):
@@ -85,26 +83,6 @@ def generate_audio(text, lang_code, voice, speed):
         buffer.seek(0)
         return buffer
 
-# Transcribe the generated audio using speech recognition
-def transcribe_audio(audio_buffer):
-    recognizer = sr.Recognizer()
-    with sr.AudioFile(audio_buffer) as source:
-        audio = recognizer.record(source)
-    try:
-        # Transcribe using Google Web Speech API (requires internet)
-        text = recognizer.recognize_google(audio)
-        return text
-    except sr.UnknownValueError:
-        return "Sorry, could not understand the audio"
-    except sr.RequestError as e:
-        return f"Request error from Google Speech Recognition service; {e}"
-
-# Romanize (convert to Romanji) if the language is Japanese
-def romanize_text(text, lang_code):
-    if lang_code == 'j':  # Japanese language code
-        return romkan.to_roma(text)  # Correct method to use
-    return text  # No need for Romanization for other languages
-
 # Generate and display the audio file
 if st.button('Generate Audio'):
     st.write("Generating speech...")
@@ -113,20 +91,10 @@ if st.button('Generate Audio'):
     # Display Audio player in the app
     st.audio(audio_buffer, format='audio/wav')
 
-    # Transcribe the generated speech to text
-    transcription = transcribe_audio(audio_buffer)
-    
-    # Romanize the transcription if it's Japanese
-    romanized_text = romanize_text(transcription, lang_code)
-    
-    # Display the transcribed and Romanized text
-    st.write("Transcribed Text: ", transcription)
-    st.write("Romanized Pronunciation: ", romanized_text)
-
     # Optional: Save the generated audio file for download
     st.download_button(
         label="Download Audio",
         data=audio_buffer,
         file_name="generated_speech.wav",
         mime="audio/wav"
-    )
+    )