Chi tiết| Tên | The 2013 KIT IWSLT Speech-to-Text Systems for German and English |
| Lĩnh vực | Tin học |
| Tác giả | Kevin Kilgour, Christian Mohr, Michael Heck, Quoc Bao Nguyen, Van Huy Nguyen, Evgeniy Shin, Igor Tseyzer, Jonas Gehring, Markus Muller, Matthias Sperber, Sebastian Stuker and Alex Waibel |
| Nhà xuất bản / Tạp chí | International Workshop on Spoken Language Translation Năm 2013 |
| Số hiệu ISSN/ISBN | |
| Tóm tắt nội dung | |
This paper describes our English Speech-to-Text (STT) systems for the 2013 IWSLT TED ASR track. The systems consist of multiple subsystems that are combinations of dif-ferent front-ends, e.g. MVDR-MFCC based and lMel based ones, GMM and NN acoustic models and different phonesets. The outputs of the subsystems are combined via confusion network combination. Decoding is done in two stages, where the systems of the second stage are adapted in an unsu-pervised manner on the combination of the first stage outputs using VTLN, MLLR, and cMLLR.
|
|
