divide and conquer, don't do tts for now

app.py

@@ -9,7 +9,7 @@ import time
 import subprocess
 from utils.stt import transcribe_audio
 from utils.translation import translate_text
-from utils.tts import generate_speech
+from utils.tts_dummy import generate_speech
 
 # Hugging Face Spaces Setup Automation
 def setup_huggingface_space():
@@ -125,7 +125,7 @@ def render_results(english_text, chinese_text, output_path):
 
 def main():
     """Main application workflow"""
-    setup_huggingface_space()  # First-run configuration checks
+    # setup_huggingface_space()  # First-run configuration checks
     configure_page()
     st.title("🎧 High-Quality Audio Translation System")
     st.markdown("Upload English Audio → Get Chinese Speech Output")

requirements.txt

@@ -11,4 +11,4 @@ phonemizer>=3.0
 espeak-ng>=1.51
 scipy>=1.11
 munch>=2.5
-git+https://github.com/hexgrad/Kokoro-82M
+# git+https://github.com/hexgrad/Kokoro-82M

utils/tts_dummy.py

@@ -0,0 +1,3 @@
+def generate_speech(text: str, language: str = "zh") -> str:
+    """Public interface for TTS generation"""
+    return "temp/outputs/dummy.wav"