Abstract: Methods and apparatus for speech recognition based on a hidden Markov model are disclosed. A disclosed method of speech recognition is based on a hidden Markov model in which words to be recognized are modeled as chains of states and trained using predefined speech data material. Known vocabulary is divided into first and second partial vocabularies where the first partial vocabulary is trained and transcribed using a whole word model and the second partial vocabulary is trained and transcribed using a phoneme-based model in order to obtain a mixed hidden Markov model. The transcriptions from the two models are stored in a single pronunciation lexicon and the mixed hidden Markov model stored in a singe search space. Apparatus are disclosed that also employ a hidden Markov model.

Abstract: A partial fragmentation bullet comprising a nose core, which forms a fragmenting portion of the partial fragmentation bullet, and a rear core, which forms a penetration portion of the partial fragmentation bullet. The nose core and the rear core are made of materials having the same level of hardness. This level of hardness is in the Brinell hardness range.