5th International Conference on Spoken Language Processing
We present constrained time alignment acoustic models based on phonetic knowledge and a speaker independent speech recognition method using our proposed models. Japanese syllable and isolated word recognition experiments show that the models have robustness to intra- and inter- speaker varieties such as acoustic diversity. Furthermore we experiment with word recognition tests under the condition such as noise environments and endpoints free matching, it reveals the feasibility of our proposed models.
Bibliographic reference. Konuma, Tomohiro / Suzuki, Tetsu / Yamada, Maki / Ohno, Yoshio / Hoshimi, Masakatsu / Niyada, Katsuyuki (1998): "Speaker independent speech recognition method using constrained time alignment near phoneme discriminative frame", In ICSLP-1998, paper 0198.