Meinhard Ullrich

Learn More
In this paper we present some experiments that have been performed while developing language models for the PHILIPS Broadcast News system. Three main issues will be discussed: construction of phrases, adaptation of remote corpora to this task, and the combination of the diierent models. Also, per-plexities on the 1997 evaluation data are reported.
The performance of the Philips system for large vocabulary continuous speech recognition has been improved signiicantly by crossword N-phone modelling, enhanced clustering of HMM-states during training, consistent handling of untrained HMM-states during decoding and a new eecient crossword N-phone M-gram decoding strategy. We report word error rate(More)
In this paper the Philips Broadcast News transcription system is described. The Broadcast News task aims at the recognition of found" speech in radio and television broadcasts without any additional side information e.g. speaking style, background conditions. The system was derived from the Philips continuous mixture density crossword HMM system, using MFCC(More)
In this paper we describe some characteristics of the acoustic modeling used in the Philips continuous-speech recognition system for the DARPA Hub-4 1997 evaluation, which are related to robustness issues. We aimed at a conceptually simple system: We trained two model sets on 70 hours of the Hub-4 training data, one for within-word and one for crossword(More)
This paper presents an extension of bottom-up state-tying towards improved handling of unseen triphones. As opposed to the usual backing-o to diphones and monophones, the current method aims at nding a triphone model that has proven to exhibit some similarity with the unseen triphone. It is based on a probabilistic mapping of unseen contexts to clusters of(More)
  • 1