I have trained a model on collected data. Now, I want to fine tune this model with my own voice data. My question is, should the language model for fine tuning contain only the transcripts of my own voice data. Or should it contain the transcripts from the previous model too?
jahir (Jahir) #1