Voice Assistant Configuration - project-owner/Peppy.doc GitHub Wiki
The Voice Assistant Configuration page can be used to manage voice models. It allows to download new model and swicth between models.
The Voice Assistant can work only if the model is selected.
The models are displayed according to the current language. The list of models will be loaded from the Vosk web site. All downloaded models will be saved to the folder specified in the text field Folder. Each model is a folder with subfolders and files.
Please check the model size before downloading. Large models need more time for downloading. It's not necessary that a large model will provide better voice recognition. Actually I was able to achieve the best results with a small model.
- Type - There is only one assistant type at the moment Voice Assistant.
- Translate Names - This property allows to translate disc and file names to English.
- Folder - The folder name where player will download voice models. The default value is /home/pi/vosk-models.
- Vosk Models - The list of Vosk Models. The models can be either downloaded (clicking on the download icon) or deleted (clicking on the delete icon). Delete action deletes files from a file system.