I just recorded a few words with simon:
Let’s take a look into /media/Hitachi/simon-xp-wav:
You can see that the file Kalifornien_erzielt_Distanz_Freundschaft_S1_2009-07-19_11-22-40.wav has been stored in that folder. Why is it being stored on my Hitachi hard drive? Take a look at the path to the training samples:
In the folder /home/liberty/.kde/share/apps/simon/model/training.data, I have two unused files:
The word Aachen is currently not part of my active vocabulary. I think that I could delete both wav files. Or could I move these two ‘Aachen’ wav files to the the folder /media/Hitachi/simon-xp-wav? Would this mean that they would become part of my speech model again (Aachen is included in the shadow dictionary)? But there are three pronunciations for Aachen available. How would simon know which one to choose?
Tags: fʀɔɪ̯ntʃaft, kalifɔrniən, training, ɛɐtsiːlt
Hi!
Apparently those samples slipped through the deletion mechanism (were they in your prompts at the time you deleted the word?).
Copying them to the directory would not add them to the model. You would have to add the transcription to the prompts file at which point simon would fail to compile the model unless you specify the phonetic transcription of the word by adding it to your active model again. At this point you would have to decide which pronunciation to use anyways.
Greetings,
Peter