论文部分内容阅读
80年代,IBM公司和其他公司推出了几个带有大型词汇表的系统,然而由于计算和准确度的限制迫使说话人说话时在两个单词之间要作一停顿。 语音领域的进步使人们对智能系统的结构有了更深入的了解,如怎样使用不完备、不准确的部分知识来解决问题。由卡内基梅隆大学研制的Sphinx Ⅲ系统不需要对每个新的说话人作任何限制就可以识别连续的语音。该系统以近乎实时的
In the 1980s, IBM and others introduced several systems with large vocabulary, however due to computational and accuracy constraints, forces the speaker to pause between two words as they speak. Advances in the field of speech have led to a deeper understanding of the structure of intelligent systems, such as how to use incomplete knowledge and inaccurate parts of knowledge to solve problems. The Sphinx III system, developed by Carnegie Mellon University, recognizes continuous speech without any restriction on each new speaker. The system is in near real time