I am kinda new to this domain and I want to be able to make an automatic speech recognition from scratch, that works offline and training my own models from scratch also.
I am struggling finding the right toolkit (kaldi, esp, deepspeech what i saw online) preferably lightweight and can do this on cpu since i dont have a cuda enabled gpu. Also any suggestions in general would be very helpful. Sorry for my english. Thank you in advance ya'll time!