Between the choice of plugins and options, configuring Naomi can be overwhelming for a new user. We would like to create a few “presets”.
“Best Privacy” would be a totally self-contained version where speech to text and text to speech all happens offline, and print_transcript and save_audio options are all turned off. Right now that would probably mean using Flite for Text to Speech and Pocketsphinx for Speech to Text.
“Best Accuracy” would probably use pocketsphinx for passive listening, and google cloud voice for active and special listening and text to speech.
“Voice Recognition developer” would turn on print_transcript and save_active_audio options, and automatically download and install all offline Speech to Text options (pocketsphinx, deepspeech, julius, kaldi).
We’d like to know what plugins and options you use and why. Please let us know by replying below.