Combining Bottleneck-BLSTM and Semi-Supervised Sparse NMF for Recognition of Conversational Speech in Highly Instationary Noise
Combining Bottleneck-BLSTM and Semi-Supervised Sparse NMF for Recognition of Conversational Speech in Highly Instationary Noise
复制标题
结合 Bottleneck-BLSTM 和半监督稀疏 NMF 来识别高度不稳定噪声中的会话语音
DOI：
--
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
Björn Schuller
中科院分区：
文献类型：
--
作者：
F. Weninger;M. Wöllmer;Björn Schuller
我们通过应用半监督稀疏非负矩阵分解（NMF）进行语音增强，并结合我们最近提出的利用双向长短期记忆（BLSTM）循环神经网络生成的瓶颈（BN）特征的前端，解决高度可变噪声中自发语音的独立自动识别。在我们的评估中，我们将 2011 年 PASCAL CHiME 挑战赛的噪声语料库和评估协议与 Buckeye 数据库结合起来，并证明了 NMF 增强和 BN-BLSTM 前端的结合在这项极具挑战性的任务中（信噪比为 -6 到 9 dB）在单词准确性方面带来了显着且一致的增益。
We address the speaker independent automatic recognition of spontaneous speech in highly variable noise by applying semisupervised sparse non-negative matrix factorization (NMF) for speech enhancement coupled with our recently proposed frontend utilizing bottleneck (BN) features generated by a bidirectional Long Short-Term Memory (BLSTM) recurrent neural network. In our evaluation, we unite the noise corpus and evaluation protocol of the 2011 PASCAL CHiME challenge with the Buckeye database, and we demonstrate that the combination of NMF enhancement and BN-BLSTM front-end introduces significant and consistent gains in word accuracy in this highly challenging task at signal-to-noise ratios from -6 to 9 dB.