Academic Journal
The 2013 KIT IWSLT Speech-to-Text Systems for German and English
| العنوان: | The 2013 KIT IWSLT Speech-to-Text Systems for German and English |
|---|---|
| المؤلفون: | Kilgour, Kevin, Mohr, Christian, Heck, Michael, Nguyen, Quoc Bao, Nguyen, Van Huy, Shin, Evgeniy, Tseyzer, Igor, Gehring, Jonas, Müller, Markus, Sperber, Matthias, Stüker, Sebastian, Waibel, Alex |
| بيانات النشر: | Association for Computational Linguistics |
| سنة النشر: | 2024 |
| المجموعة: | KITopen (Karlsruhe Institute of Technologie) |
| مصطلحات موضوعية: | ddc:004, DATA processing & computer science, info:eu-repo/classification/ddc/004 |
| الوصف: | 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 different front-ends, e.g. MVDR-MFCC based and lMel based ones, GMM and NN acoustic models and different phone sets. 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 unsupervised manner on the combination of the first stage outputs using VTLN, MLLR, and cMLLR. |
| نوع الوثيقة: | article in journal/newspaper conference object |
| وصف الملف: | application/pdf |
| اللغة: | English |
| Relation: | https://publikationen.bibliothek.kit.edu/1000166322; https://publikationen.bibliothek.kit.edu/1000166322/152200537; https://doi.org/10.5445/IR/1000166322 |
| DOI: | 10.5445/IR/1000166322 |
| الاتاحة: | https://publikationen.bibliothek.kit.edu/1000166322 https://publikationen.bibliothek.kit.edu/1000166322/152200537 https://doi.org/10.5445/IR/1000166322 |
| Rights: | https://creativecommons.org/licenses/by/4.0/deed.de ; info:eu-repo/semantics/openAccess |
| رقم الانضمام: | edsbas.3978BC0F |
| قاعدة البيانات: | BASE |
| DOI: | 10.5445/IR/1000166322 |
|---|