Title :
Towards universal speech recognition
Author :
Wang, Zhirong ; Topkara, Umut ; Schultz, Tanja ; Waibel, Alex
Author_Institution :
Interactive Syst. Labs., Carnegie Mellon Univ., Pittsburgh, PA, USA
Abstract :
The increasing interest in multilingual applications like speech-to-speech translation systems is accompanied by the need for speech recognition front-ends in many languages that can also handle multiple input languages at the same time. We describe a universal speech recognition system that fulfills such needs. It is trained by sharing speech and text data across languages and thus reduces the number of parameters and overhead significantly at the cost of only slight accuracy loss. The final recognizer eases the burden of maintaining several monolingual engines, makes dedicated language identification obsolete and allows for code-switching within an utterance. To achieve these goals we developed new methods for constructing multilingual acoustic models and multilingual n-gram language models.
Keywords :
computational linguistics; language translation; natural language interfaces; speech recognition; speech-based user interfaces; code-switching; dedicated language identification; monolingual engines; multilingual acoustic models; multilingual applications; multilingual n-gram language models; multiple input languages; overhead; speech-to-speech translation; text data; universal speech recognition; Application software; Costs; Engines; Interactive systems; Laboratories; Natural languages; Speech processing; Speech recognition; Switches; Text recognition;
Conference_Titel :
Multimodal Interfaces, 2002. Proceedings. Fourth IEEE International Conference on
Print_ISBN :
0-7695-1834-6
DOI :
10.1109/ICMI.2002.1167001