name: llama-lava-vision-speech-assistant runtime: python3.10 requirements: - transformers>=4.30.0 - diffusers>=0.18.0 - gradio>=3.30.0 - opencv-python>=4.7.0 - soundfile>=0.11.0 - speechrecognition>=3.8.1 # - vosk (If you are using Vosk for speech recognition) - pillow>=9.4.0 # ... (Add any other required libraries)