Speech Translation Enhanced ASR for European Parliament Speeches - On the Influence of ASR Performance on Speech Translation

Abstract

In this paper we describe our work in coupling automatic speech recognition (ASR) and machine translation (MT) in a speech translation enhanced automatic speech recognition (STE-ASR) framework for transcribing and translating European Parliament speeches. We demonstrate the influence of the quality of the ASR component on the MT performance, by comparing a series of WERs with the corresponding automatic translation scores. By porting an STE-ASR framework to the task at hand, we show how the word errors for transcribing English and Spanish speeches can be lowered by 3.0% and 4.8% relative, respectively.

Simultaneous translation of lectures and speeches

“…From results obtained within TC-STAR and from our studies in Stüker et al (2007) it can be seen that there is an negative almost linear correlation between the WER and MT quality.…”

“…As expected, the performance on ASR results is worse compared to that on manually transcripts. It should be noted that from results obtained within TC-STAR and from studies done by ourselves in Stüker et al (2007) it can be seen that there is an negative almost linear correlation between the WER and the MT quality.…”

“…We used ASR hypotheses as well as reference transcripts for the experiments, whereas the Spanish hypotheses were generated with a system trained within TC-STAR on Parliament plenary sessions (Stüker et al 2007;Paulik and Waibel 2008). The case-insensitive WER was 8.4%.…”