The TUM+TUT+KUL Approach to the 2nd CHiME Challenge: Multi-Stream ASR Exploiting BLSTM Networks and Sparse NMF
The TUM+TUT+KUL Approach to the 2nd CHiME Challenge: Multi-Stream ASR Exploiting BLSTM Networks and Sparse NMF
复制标题
TUM TUT KUL 应对第二届 CHiME 挑战赛的方法：利用 BLSTM 网络和稀疏 NMF 的多流 ASR
DOI：
--
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
T. Virtanen
中科院分区：
文献类型：
--
作者：
Jürgen T. Geiger;F. Weninger;Antti Hurmalainen;J. Gemmeke;M. Wöllmer;Björn Schuller;G. Rigoll;T. Virtanen
我们将共同为第二届CHiME语音分离和识别挑战赛做出贡献。我们的系统结合了语音增强的监督稀疏非负矩阵分解（NMF）与多流语音识别系统。除了传统的MFCC HMM识别器之外，还将双向长短期记忆递归神经网络（BLSTM-RNN）和非负稀疏分类（NSC）的预测集成到三流识别器中。在小词汇量和中等词汇量识别任务的挑战进行了实验。在挑战基线上的持续改进证明了所提出的系统的有效性，在小词汇量任务中的平均单词准确率为92.8%，在中等词汇量任务中的平均单词错误率为41.42%。
We present our joint contribution to the 2nd CHiME Speech Separation and Recognition Challenge. Our system combines speech enhancement by supervised sparse non-negative matrix factorisation (NMF) with a multi-stream speech recognition system. In addition to a conventional MFCC HMM recogniser, predictions by a bidirectional Long Short-Term Memory recurrent neural network (BLSTM-RNN) and from non-negative sparse classification (NSC) are integrated into a triple-stream recogniser. Experiments are carried out on the small vocabulary and the medium vocabulary recognition tasks of the Challenge. Consistent improvements over the Challenge baselines demonstrate the efficacy of the proposed system, resulting in an average word accuracy of 92.8% in the small vocabulary task and an average word error rate of 41.42% in the medium vocabulary task.