Zobrazeno 1 - 1
of 1
pro vyhledávání: '"程永任"'
Autor:
Yung-Jen Cheng, 程永任
96
The traditional acoustic model training is based on Maximum Likelihood (ML). This training method maximizes the posterior probability of transcription in the training corpus, but cannot guarantee that the incorrect observations do not obtain
The traditional acoustic model training is based on Maximum Likelihood (ML). This training method maximizes the posterior probability of transcription in the training corpus, but cannot guarantee that the incorrect observations do not obtain
Externí odkaz:
http://ndltd.ncl.edu.tw/handle/99061010271406424560
