Investigating recognition of children's speech
Investigating recognition of children's speech
复制标题
调查儿童言语识别
DOI：
10.1109/icassp.2003.1202313
复制
发表时间：
2003
期刊：
影响因子：
--
通讯作者：
M. Gerosa
中科院分区：
文献类型：
--
作者：
D. Giuliani;M. Gerosa
通过考虑电话识别任务来研究儿童语音识别。通过利用两个意大利语语音数据库，训练了两个基线系统，一个用于儿童，一个用于成人。在匹配条件下，使用同一人群的数据进行训练和识别，儿童和成人的手机识别准确率分别为77.30%和79.43%。研究发现，对于许多儿童来说，识别结果与成人一样好。然而，观察到儿童的电话识别准确度差异高于成人。还研究了匹配和不匹配的训练和测试条件下的声道长度标准化。对于成人和儿童，观察到相对于基线系统的性能改善。
Recognition of children's speech was investigated by considering a phone recognition task. Two baseline systems were trained, one for children and one for adults, by exploiting two Italian speech databases. Under matching conditions, for training and recognition performed with data from the same population group, the phone recognition accuracy was 77.30% and 79.43% for children and adults, respectively. It was found that, for many children, recognition results were as good as for adults. However, a higher variability in phone recognition accuracy across speakers was observed for children than for adults. Vocal tract length normalization, under matched and mismatched training and testing conditions, was also investigated. For both adults and children, a performance improvement, with respect to the baseline systems, was observed.