Adaptive ML-weighting in multi-band recombination of Gaussian mixture ASR
Adaptive ML-weighting in multi-band recombination of Gaussian mixture ASR
复制标题
高斯混合 ASR 多频带重组中的自适应 ML 加权
DOI：
10.1109/icassp.2001.940816
复制
发表时间：
2001
期刊：
影响因子：
--
通讯作者：
A. Morris
中科院分区：
文献类型：
--
作者：
Astrid Hagen;H. Bourlard;A. Morris
多频段语音识别在带限噪声中具有较强的识别能力，在重组过程中，可以对可靠性较低的噪声频段的识别器给予较小的权重。通常很难准确地决定哪些频段可以被认为是可靠的，哪些频段由于噪声破坏而不那么可靠。我们研究了一种最大似然(ML)方法来自适应多频带系统的组合权重。高斯混合模型参数保持不变，而组合权值迭代更新以最大化数据似然。将无监督的离线和在线权重自适应与使用相同的权重进行比较，并在已知噪声频段的情况下使用“作弊”权重，以及与全频段系统进行比较。初步测试表明，两种最大似然加权策略对带限噪声都有较好的鲁棒性。
Multi-band speech recognition is powerful in band-limited noise, when the recognizer of the noisy band, which is less reliable, can be given less weight in the recombination process. An accurate decision on which bands can be considered as reliable and which bands are less reliable due to corruption by noise is usually hard to take. We investigate a maximum-likelihood (ML) approach to adapting the combination weights of a multi-band system. The Gaussian mixture model parameters are kept constant, while the combination weights are iteratively updated to maximize the data likelihood. Unsupervised offline and online weights adaptation are compared to the use of equal weights, and 'cheating' weights where the noisy band is known, as well as to the fullband system. Initial tests show that both ML-weighting strategies show a robustness gain on band-limited noise.