Multi-Style Reverberation Models and Efficient Model Adaptation for Robust Distant-Talking Speech Recognition with REMOS

Konferenz: Sprachkommunikation 2010 - 9. ITG-Fachtagung
06.10.2010 - 08.10.2010 in Bochum, Deutschland

Tagungsband: Sprachkommunikation 2010

Seiten: 4Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Maas, Roland; Sehr, Armin; Kellermann, Walter (Multimedia Communications and Signal Processing, University of Erlangen-Nuremberg, Cauerstr. 7, 91058 Erlangen, Germany)

To further increase the flexibility of the REMOS (REverberation MOdeling for Speech recognition) concept for distant-talking speech recognition, multi-style reverberation models (RVMs) trained on data from different rooms as well as simplified RVMs are analyzed in this contribution. If the multi-style probability density functions (pdfs) used for score calculation are adapted to the reverberation conditions of the current room, a remarkable improvement in recognition performance can be achieved. Evaluations of a very efficient reverberation model adaptation scheme by connected digit recognition experiments show that REMOS can be adjusted to different reverberation conditions with minimal effort.