A Comparison of Hybrid and End-to-End Models for Syllable Recognition

  • This paper presents a comparison of a traditional hybrid speech recognition system (kaldi using WFST and TDNN with lattice-free MMI) and a lexicon-free end-to-end (TensorFlow implementation of multi-layer LSTM with CTC training) models for German syllable recognition on the Verbmobil corpus. The results show that explicitly modeling prior knowledge is still valuable in building recognition systems. With a strong language model (LM) based on syllables, the structured approach significantly outperforms the end-to-end model. The best word error rate (WER) regarding syllables was achieved using kaldi with a 4-gram LM, modeling all syllables observed in the training set. It achieved 10.0% WER w.r.t. the syllables, compared to the end-to-end approach where the best WER was 27.53%. The work presented here has implications for building future recognition systems that operate independent of a large vocabulary, as typically used in a tasks such as recognition of syllabic or agglutinative languages, out-of-vocabulary techniques, keyword searchThis paper presents a comparison of a traditional hybrid speech recognition system (kaldi using WFST and TDNN with lattice-free MMI) and a lexicon-free end-to-end (TensorFlow implementation of multi-layer LSTM with CTC training) models for German syllable recognition on the Verbmobil corpus. The results show that explicitly modeling prior knowledge is still valuable in building recognition systems. With a strong language model (LM) based on syllables, the structured approach significantly outperforms the end-to-end model. The best word error rate (WER) regarding syllables was achieved using kaldi with a 4-gram LM, modeling all syllables observed in the training set. It achieved 10.0% WER w.r.t. the syllables, compared to the end-to-end approach where the best WER was 27.53%. The work presented here has implications for building future recognition systems that operate independent of a large vocabulary, as typically used in a tasks such as recognition of syllabic or agglutinative languages, out-of-vocabulary techniques, keyword search indexing and medical speech processing.show moreshow less

Export metadata

Additional Services

Search Google Scholar
Metadaten
Author:Korbinian Riedhammer, Sebastian P. BayerlORCiD
DOI:https://doi.org/10.1007/978-3-030-27947-9_30
ArXiv Id:http://arxiv.org/abs/arXiv:1909.12232
Document Type:conference proceeding (article)
Language:English
Date of first Publication:2019/09/19
Reviewed:Begutachtet/Reviewed
Release Date:2024/07/05
Tag:speech recognition, language model, CTC, end-2-end, syllables
Pagenumber:8
Konferenzangabe:International Conference on Text, Speech and Dialogue (TSD)
institutes:Fakultät Informatik
Licence (German):Keine Lizenz - Deutsches Urheberrecht gilt
Verstanden ✔
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.