Recent improvements of the RWTH large vocabulary speech recognition system on spontaneous speech

Author

Sixtus, Achim ; Molau, Sirko ; Kanthak, Stephan ; Schluter, Ralf ; Ney, Hermann

Author_Institution

Luhrstuhl fur Inf., Tech. Hochschule Aachen, Germany

Volume

3

fYear

2000

fDate

2000

Firstpage

1671

Abstract

The paper presents recent improvements of the RWTH large vocabulary continuous speech recognition system (LVCSR). In particular, we report on the integration of across-word models into the first recognition pass, and describe better algorithms for fast vocal tract normalization (VTN). We focus both on improvements in word error rate and how to speed up the recognizer with only minimal loss of recognition accuracy. Implementation details and experimental results are given for the VerbMobil task, a German spontaneous speech corpus. The 25.0% word error rate (WER) of our within-word baseline system was reduced to 21.4% with VTN and across-word models. Decreasing the real-time factor (RTF) by up to 85% resulted in only a small degradation in recognition performance of 2% relative on average

Keywords

speech recognition; vocabulary; German spontaneous speech corpus; RWTH large vocabulary continuous speech recognition system; VerbMobil task; across-word models; fast vocal tract normalization; real-time factor; recognition accuracy; spontaneous speech; within-word baseline system; word error rate; Decoding; Degradation; Error analysis; Hidden Markov models; Real time systems; Speech recognition; Testing; Vectors; Viterbi algorithm; Vocabulary;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2000. ICASSP '00. Proceedings. 2000 IEEE International Conference on

Conference_Location

Istanbul

ISSN

1520-6149

Print_ISBN

0-7803-6293-4

Type

conf

DOI

10.1109/ICASSP.2000.862071

Filename

862071