摘要
HMM的观测概率能否很好描述模型的实际分布对识别性能有很大的影响。为了比较汉语三音子模型在不同观测概率情况下的差异 ,本文构造了三种不同模型 ,及其训练和识别算法。通过从多方面对这三个模型进行比较 ,得出结论 ,为今后汉语三音子模型观测概率的选择提供依据。
It has great impact on the recognition performance how well HMM observation density can describe the actual distribution. To compare triphone models under different observation densities in Chinese language, three models and their respective algorithms of training and recognizing are constructed. By comparison of these three models in terms of different aspects, a conclusion is drawn, which can be the basis for the future selection of triphone observation density in Chinese language.
出处
《中文信息学报》
CSCD
北大核心
2003年第3期47-52,共6页
Journal of Chinese Information Processing
基金
国家 8 63高技术项目 ( 863- 30 6-ZD0 3- 0 2 - 1)
985重大项目 ( 985校 - 2 2 -攻关 - 0 6)
关键词
人工智能
机器翻译
语音识别
三音子模型
观测概率
artificial intelligence
machine translation
speech recognition
triphone model
observation density
