Utterance-level boosting of HMM speech recognizers
Utterance-level boosting of HMM speech recognizers
复制标题
HMM 语音识别器的话语级别提升
DOI：
10.1109/icassp.2002.5743666
复制
发表时间：
2002
期刊：
影响因子：
--
通讯作者：
C. Meyer
中科院分区：
文献类型：
--
作者：
C. Meyer
我们提出了一种提高HMM语音识别者的话语级别的方法。标准的adaboost.m2算法用于计算说服的训练权重，这些话语用于后续声学模型的最大似然（ML）训练。认识到，这些模型的得分是线性合并的。我们在大型词汇隔离单词识别任务上评估了算法。与ML基线模型相比，即使对于高密度的声学模型，也获得了显着的性能改进（相对高达9％）。特别是，结合增强模型的分数优于组合来自不同ML基线模型的得分。
We propose an utterance-level approach to boosting HMM speech recognizers. The standard AdaBoost.M2 algorithm is applied to calculate training weights for utterances, which are used in maximum likelihood (ML) training of subsequent acoustic models. In recognition, scores of these models are linearly combined. We evaluate our algorithm on a large vocabulary isolated word recognition task. Significant performance improvements (up to 9% relative) are obtained compared to the ML baseline model, even for high density acoustic models. In particular, combining scores from boosted models outperforms combining scores from different ML baseline models.