Actually, I have watched a lot of tutoriel and I’ve tried a lot of ways. But i failed to integrate voice in my rasa bot. It seem that the new version of python doesn’t work with rasa. Any way my goal is to create a bot which can communicate with user via text or voice.
You can choose your own voice models in the credentials YAML file for voice_connector, I have used my own trained nvidia models and also already trained silero-models from torch hub, better than Mozilla’s ones, at least for spanish language (nvidia with CUDA try only if you have or you want to train your own voice models because it creates really big containers and requires specific nvidia docker runtime).