1 paper
Haris Bin Zia, Agha Ali Raza, Awais Athar
State-of-the-art speech recognition systems rely heavily on three basic components: an acoustic model, a pronunciation lexicon and a language model. To build these components, a re…