1-2hit |
Keiji YASUDA Hirofumi YAMAMOTO Eiichiro SUMITA
For statistical language model training, target domain matched corpora are required. However, training corpora sometimes include both target domain matched and unmatched sentences. In such a case, training set selection is effective for both reducing model size and improving model performance. In this paper, training set selection method for statistical language model training is described. The method provides two advantages for training a language model. One is its capacity to improve the language model performance, and the other is its capacity to reduce computational loads for the language model. The method has four steps. 1) Sentence clustering is applied to all available corpora. 2) Language models are trained on each cluster. 3) Perplexity on the development set is calculated using the language models. 4) For the final language model training, we use the clusters whose language models yield low perplexities. The experimental results indicate that the language model trained on the data selected by our method gives lower perplexity on an open test set than a language model trained on all available corpora.
We present the 10-bit current driver LSI with 2-set current digital-to-analog converters (DACs) and output channel current sample and hold (S/H) circuits for large-size and high-resolution active matrix organic light emitting diode (AMOLED) display applications. This current driver LSI has 300 output channels and the output current ranges from 0 µA to 290 µA. The maximum output current level can be controlled by 2-bit control signals because the maximum output current level depends on display size and resolution. The chip was fabricated using 0.65µm BiCMOS process and characterized. The chip size is 16.8 mm3.6 mm. Experimental results show that the output current DNL is less than 0.4 LSB and that INL is less than 1.5 LSB. This is good enough to apply 15.5 inch WXGA (1280RGB768) AMOLED displays.