Scientific Work

Dresden University of Technology

Hofmann, Michael: Prosodic and Phonetic Segmentation and Annotation of German and English Speech Corpora. Master's Thesis. Dresden University of Technology, 2006.

Hofmann, Michael: Optimierung einer trainingsbasierten Prosodiegenerierung. Studienarbeit, Technische Universität Dresdenn, 2005.

Publications

Hofmann, Michael and Oliver Jokisch: Optimization of MFNs for signal-based phrase break prediction. In SpeechProsody '06, Dresden, Germany, 2006.

Hofmann, Michael and Oliver Jokisch: OpenVOC - Open Platform for Multilingual Vocabulary Training Integrating Speech Technology Components, In Proceedings of the 10th International Workshop Speech and Computer (SPECOM), Patras, Greece, 2005.

Hofmann, Michael and Boris Lobanov: Vocabulary Training Program Using TTS and Speech Recognition Technologies. In Proceedings of the 2nd Baltic Conference on Human Language Technologies (HLT), Tallin, Estonia, 2005.

Jokisch, Oliver und Michael Hofmann: Optimierung einer trainingsbasierten Prosodiegenerierung für Sprachsynthese. In 15. Konferenz Elektronische Sprachsignalverarbeitung (ESSV), Cottbus, Deutschland, 2004.

Jokisch, Oliver and Michael Hofmann: Evolutionary Optimization of an Adaptive Prosody Model. In Proceedings of the ICSLP '04, pages 797-800, Jeju Island, South Korea, 2004.