gradio
gTTS
speechrecognition
sounddevice
scipy
ffmpeg-python