研究 Research
Emotion Profile Refinery for Speech Emotion Classification
Human emotions are inherently ambiguous and impure. When
designing systems to anticipate human emotions based on
speech, the lack of emotional purity must be considered.
- 2020/8/12
- arXiv preprint arXiv:2008.05259
- Shuiyang Mao, PC Ching, Tan Lee
Automatic Detection of Phonological Errors in Child Speech Using Siamese Recurrent Autoencoder
Speech sound disorder (SSD) refers to the developmental disorder in which children encounter persistent difficulties in correctly pronouncing words.
- 2020/8/7
- arXiv preprint arXiv:2008.03193
- Si-Ioi Ng, Tan Lee
日本名古屋。1993
基於神經網絡的粵語聲調分類器
李丹教授；程伯中教授；陳麗雲教授發表
聲調識別無疑是中文語音識別問題中的一個重要組成部分，特別是對於眾所周知的聲調豐富的粵語而言。
使用包含 234 個不同音節的大詞彙表，單說話人和多說話人情況下的系統性能分別為 89% 和 87%。
對於每個特定的語音單元，構建一個完全連接的循環神經網絡，以便通過神經元激活狀態的特定時間模式同時表示靜態和動態語音特徵。