摘要 该文提出了一种基于音素网格的置信度计算方法。与传统的基于整个声学模型的置信度不同的是，这种方法在解码器生成的音素网格上计算关键词的置信度，从而具有更好的拒识能力。另外，针对两种置信度取值范围的不同，该文采用权重因子的方法综合利用两种置信度，取得了较好的效果。在自然对话的电话数据测试中，与传统的置信度计算方式相比，混和置信度的FOM(Figure Of Merit)值相对提高了17.0%。
Abstract：Phoneme lattice based Confidence Measure (CM) is proposed in this paper. It makes use of phoneme lattices generated by a phoneme recognizer. Acoustic Model (AM) based CM is also introduced. For a decoded speech frame aligned to an HMM state, the CM based on AM is calculated. These two confidence measures are combined using a weighting factor to obtain a hybrid CM as they had different dynamic scales. On spontaneous conversational telephone database, the Figure Of Merit (FOM) achieves 17.0% relative improvement comparing to AM based CM.
2007, Vol. 29 
