The aim of works described in this article is to elaborate and experimentally evaluate a consistent method of Language Model (LM) construction for the sake of Polish speech recognition. In the proposed method we tried to take into account the features and specific problems experienced in practical applications of speech recognition in the Polish language, reach inflection, a loose word order and the tendency for short word deletion. The LM is created in five stages. Each successive stage takes the model prepared at the previous stage and modifies or extends it so as to improve its properties. At the first stage, typical methods of LM smoothing are used to create the initial model. Four most frequently used methods of LM construction are here. At the second stage the model is extended in order to take into account words indirectly co-occurring in the corpus. At the next stage, LM modifications are aimed at reduction of short word deletion errors, which occur frequently in Polish speech recognition. The fourth stage extends the model by insertion of words that were not observed in the corpus. Finally the model is modified so as to assure highly accurate recognition of very important utterances. The performance of the methods applied is tested in four language domains.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The currently dominant speech recognition technology, hidden Mar-kov modeling, has long been criticized for its simplistic assumptions about speech, and especially for the naive Bayes combination rule inherent in it. Many sophisticated alternative models have been suggested over the last decade. These, however, have demonstrated only modest improvements and brought no paradigm shift in technology. The goal of this paper is to examine why HMM performs so well in spite of its incorrect bias due to the naive Bayes assumption. To do this we create an algorithmic framework that allows us to experiment with alternative combination schemes and helps us understand the factors that influence recognition performance. From the findings we argue that the bias peculiar to the naive Bayes rule is not really detrimental to phoneme classification performance. Furthermore, it ensures consistent behavior in outlier modeling, allowing efficient management of insertion and deletion errors.
3
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
A packet buffer limited to a fixed number of packets (regardless of their lengths) is considered. The buffer is described as a finite FIFO queuing system fed by a Markovian Arrival Process (MAP) with service times forming a Semi-Markov (SM) process (MAP/SM/1/b in Kendall's notation). Such assumptions allow us to obtain new analytical results for the queuing characteristics of the buffer. In the paper, the following are considered: the time to fill the buffer, the local loss intensity, the loss ratio, and the total number of losses in a given time interval. Predictions of the proposed model are much closer to the trace-driven simulation results compared with the prediction of the MAP/G/1/b model.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.