Tonotopic multi-layered perceptron: a neural network for learning long-term temporal features for speech recognition
Tonotopic multi-layered perceptron: a neural network for learning long-term temporal features for speech recognition
复制标题
Tonotopic 多层感知器：用于学习语音识别的长期时间特征的神经网络
DOI：
10.1109/icassp.2005.1415271
复制
发表时间：
2005
期刊：
影响因子：
--
通讯作者：
N. Morgan
中科院分区：
文献类型：
--
作者：
Barry Y. Chen;Q. Zhu;N. Morgan
我们一直在通过使用多层感知器（MLP）捕获长期（/spl sim/500ms）时间信息来降低会话电话语音（CTS）任务的单词错误率（WER）。在本文中，我们实验了一种称为tonotopic MLP （TMLP）的MLP架构，它包含两个隐藏层。其中第一个是拓扑组织的：对于每个关键波段，有一组不相交的隐藏单元，它们使用长期能量轨迹作为输入。因此，这些隐藏单元的每个子集都学习区分单带能量轨迹模式。其余的层完全连接到它们的输入。当与中期（/spl sim/100ms） MLP系统结合使用以增强标准PLP功能时，TMLP将2001年Nist Hub-5 CTS评估集（Eval2001）上的WER相对降低了8.87%。与以前的方法相比，我们显示出一些实际的优点。我们还报告了一系列实验的结果，以确定关于该任务和体系结构的训练模式数量的隐藏层大小和总参数的最佳范围。
We have been reducing word error rates (WER) on conversational telephone speech (CTS) tasks by capturing long-term (/spl sim/500ms) temporal information using multilayered perceptrons (MLP). In this paper we experiment with an MLP architecture called tonotopic MLP (TMLP), incorporating two hidden layers. The first of these is tonotopically organized: for each critical band, there is a disjoint set of hidden units that use the long-term energy trajectory as the input. Thus, each of these subsets of hidden units learns to discriminate single band energy trajectory patterns. The rest of the layers are fully connected to their inputs. When used in combination with an intermediate-term (/spl sim/100ms) MLP system to augment standard PLP features, the TMLP reduces the WER on the 2001 Nist Hub-5 CTS evaluation set (Eval2001) by 8.87% relative. We show some practical advantages over our previous methods. We also report results from a series of experiments to determine the best ranges of hidden layer sizes and total parameters with respect to the number of training patterns for this task and architecture.