I have dataset created in Kaldi format, so I want to build powerful ASR for uzbek language . I searched several ways to fine tuning and train , but I cannot find enough information. I want to learn to fine tuning my own dataset to Vosk toolkit using Kaldi.
and there is other issue , too which is lexicon dictionary. How to build lexicon dictionary.
and if anyone has guide to train dataset , please share it .
I have dataset created in Kaldi format, so I want to build powerful ASR for uzbek language . I searched several ways to fine tuning and train , but I cannot find enough information. I want to learn to fine tuning my own dataset to Vosk toolkit using Kaldi.
and there is other issue , too which is lexicon dictionary. How to build lexicon dictionary. and if anyone has guide to train dataset , please share it .
Thanks your respose.