Cross Domain Automatic Transcription on the TC-STAR EPPS Corpus

Gollan, C.; Bisani, Maximilian; Kanthak, Stephan; Schlüter, Ralf; Ney, Hermann

doi:10.1109/icassp.2005.1415241

Cited by 30 publications

(16 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore the automatically produced transcriptions will be consistent with the pronunciations taken over from the existing dictionary. Gollan et al (2005) as well as Lööf et al (2006) report on systems that have been ported to a new domain using this technique. A grapheme-tophoneme converter that was developed independently will generally deviate in terms of the phoneme inventories and transcription conventions used.…”

Section: Lexicon Augmentationmentioning

confidence: 99%

Joint-sequence models for grapheme-to-phoneme conversion

Bisani¹,

Ney²

2008

Speech Communication

504

471

View full text Add to dashboard Cite

Section: Lexicon Augmentationmentioning

confidence: 99%

Joint-sequence models for grapheme-to-phoneme conversion

Bisani¹,

Ney²

2008

Speech Communication

504

471

View full text Add to dashboard Cite

“…The EPPS corpora were built within the European project Technology and Corpora for Speech to Speech Translation (TC-STAR) [4,12]. The corpus statistics are given in Table 1.…”

Section: Epps English Corpusmentioning

confidence: 99%

Gammatone Features and Feature Combination for Large Vocabulary Speech Recognition

Schlüter

Bezrukov

Wagner

et al. 2007

2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07

Self Cite

136

View full text Add to dashboard Cite

In this work, an acoustic feature set based on a Gammatone filterbank is introduced for large vocabulary speech recognition. The Gammatone features presented here lead to competitive results on the EPPS English task, and considerable improvements were obtained by subsequent combination to a number of standard acoustic features, i.e. MFCC, PLP, MF-PLP, and VTLN plus voicedness. Best results were obtained when combining Gammatone features to all other features using weighted ROVER, resulting in a relative improvement of about 12% in word error rate compared to the best single feature system. We also found that ROVER gives better results for feature combination than both log-linear model combination and LDA.

show abstract

“…Within the TC-STAR project, the data consists mainly of texts of the public European Parliament Plenary Sessions (EPPS), held from 1996 to 2004, in English and Spanish, and the corresponding minutes edited by the European Parliament also known as the Final Text Editions [6]. The training portion of the data dates from April 1996 through October 16, 2004.…”

Section: Task and Data Descriptionmentioning

confidence: 99%

Investigating translation of Parliament speeches

Déchelotte

Schwenk

Gauvain

et al. 2005

IEEE Workshop on Automatic Speech Recognition and Understanding, 2005.

View full text Add to dashboard Cite

This paper reports on recent experiments for speech to text (STT) translation of European Parliamentary speeches. A Spanish speech to English text translation system has been built using data from the TC-STAR European project. The speech recognizer is a state-of-the-art multipass system trained for the Spanish EPPS task and the statistical translation system relies on the IBM-4 model. First, MT results are compared using manual transcriptions and 1-best ASR hypotheses with different word error rates. Then, an n-best interface between the ASR and MT components is investigated to improve the STT process. Derivation of the fundamental equation for machine translation suggests that the source language model is not necessary for STT. This was investigated by using weak source language models and by n-best rescoring adding the acoustic model score only. A significant loss in the BLEU score was observed suggesting that the source language model is needed given the insufficiencies of the translation model. Adding the source language model score in the n-best rescoring process recovers the loss and slightly improves the BLEU score over the 1-best ASR hypothesis. The system achieves a BLEU score of 37.3 with an ASR word error rate of 10% and a BLEU score of 40.5 using the manual transcripts.

show abstract

Cross Domain Automatic Transcription on the TC-STAR EPPS Corpus

Cited by 30 publications

References 7 publications

Joint-sequence models for grapheme-to-phoneme conversion

Joint-sequence models for grapheme-to-phoneme conversion

Gammatone Features and Feature Combination for Large Vocabulary Speech Recognition

Investigating translation of Parliament speeches

Contact Info

Product

Resources

About