A stream-weight optimization method for multi-stream HMMs based on likelihood value normalization
A stream-weight optimization method for multi-stream HMMs based on likelihood value normalization
复制标题
基于似然值归一化的多流HMM流权重优化方法
DOI：
10.1109/icassp.2005.1415152
复制
发表时间：
2005
期刊：
影响因子：
--
通讯作者：
S. Furui
中科院分区：
文献类型：
--
作者：
S. Tamura;K. Iwano;S. Furui
在音视频语音识别领域，多流隐马尔可夫模型被广泛应用，如何在小数据集上自动合理地确定流权重因子成为一个重要的研究课题。提出了一种基于输出似然归一化准则的流权重优化方法。该方法基于似然比最大化原理，通过调整流权值，使所有HMM的对数似然均值相等，从而在大规模优化数据集上取得了显著的改进。新的方法进行评估，使用日本连接数字语音记录在现实世界的环境。使用10秒语音数据进行流权重优化，与优化前的结果相比，实现了10%的绝对准确性提高。通过另外应用MLLR（最大似然线性回归）自适应，获得了23%的改进，超过了仅音频方案。
In the field of audio-visual speech recognition, multi-stream HMM are widely used, thus how to automatically and properly determine stream weight factors using a small data set becomes an important research issue. This paper proposes a new stream-weight optimization method based on an output likelihood normalization criterion. In this method, the stream weights are adjusted to equalize the mean values of log likelihood for all HMM based on likelihood-ratio maximization which achieved significant improvement by using a large optimization data set. The new method is evaluated using Japanese connected digit speech recorded in real-world environments. Using 10 seconds speech data for stream-weight optimization, a 10% absolute accuracy improvement is achieved compared to the result before optimization. By additionally applying the MLLR (maximum likelihood linear regression) adaptation, a 23% improvement is obtained over the audio-only scheme.
影响因子：
4.3
作者：
LEGGETTER, CJ;WOODLAND, PC
通讯作者：
WOODLAND, PC