Inhalt des Dokuments
zur Navigation
Work
- © Copyright??
Research Field
quality, speech technology
Research Topics
quality and quality dimensions of synthetic speech
instrumental quality prediction of synthetic speech
Biography
Florian Hinterleitner studied Communication and Computer Sciene at the Technical University Berlin. In 2010 he completed his Magister Thesis "Signalbased Quality Prediction of Synthetic Speech". He is currently working as a research assistant at the Quality and Usability Lab of Telekom Innovation Laboratories, TU-Berlin in the domain of quality prediction of synthetic speech.
Projects
Teaching
Speech Communication (since winter semester 2010/2011)
Publications
| Zitatschlüssel | norrenbrock2015a |
|---|---|
| Autor | Norrenbrock, Christoph Ritter and Hinterleitner, Florian and Heute, Ulrich and Möller, Sebastian |
| Seiten | 17–35 |
| Jahr | 2015 |
| ISSN | 0167-6393 |
| DOI | 10.1016/j.specom.2014.06.003 |
| Adresse | New York, USA |
| Journal | Speech Communication |
| Jahrgang | 66 |
| Monat | feb |
| Notiz | print/online |
| Verlag | Elsevier |
| Wie herausgegeben | full |
| Zusammenfassung | Instrumental speech-quality prediction for text-to-speech signals is explored in a twofold manner. First, the perceptual quality space of TTS is structured by means of three perceptual quality dimensions which are derived from multiple auditory tests. Second, quality-prediction models are evaluated for each dimension using prosodic and MFCC-based measurands. Linear and nonlinear model types are compared under cross-validation restrictions, giving detailed insight into model-generalizability aspects. Perceptually regularized properties, denoted as quality elements, are introduced in order to encode the quality-indicative effect of individual signal characteristics. These elements integrate a perceptual model reference which is derived in a semi-supervised fashion from natural and synthetic speech. The results highlight the feasibility of instrumental quality prediction for TTS signals provided that broad training material is employed. High prediction accuracy, however, requires nonlinear model structures. |