The Speech Understanding and Dialog System Evar

Niemann, Heinrich; Brietzmann, Astrid; Mühlfeld, Reinhard; Regel, P.; Schukat, G

doi:10.1007/978-3-642-82447-0_10

Search citation statements

Order By: Relevance

Paper Sections

Select...

Wissen üBer Tasks1

Data1

Citation Types

Supporting

Mentioning

Contrasting

Unclassified

Year Published

1985

2020

Publication Types

Select...

Book6

Relationship

Self Cite0

Independent6

Authors

Journals

Cited by 20 publications

(2 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Oft sind die Domänen-Tasks selbst sehr einfach, zum Beispiel Suchen nach einer Zugverbindung [44], Buchen nach Restaurants [77], Erzeugung von Suchanfragen an ein automatisiertes FAQ-System [24,66] -es gibt viele weitere Anwendungsszenarien. Charakteristisch für alle ist, dass die Interaktion Werte für Parameter einer domänenspezifischen Funktion "einsammelt".…”

Section: Wissen üBer Tasksunclassified

20 Assistenzsysteme

2020

Handbuch Der Künstlichen Intelligenz

View full text Add to dashboard Cite

show abstract

Section: Wissen üBer Tasksunclassified

20 Assistenzsysteme

2020

Handbuch Der Künstlichen Intelligenz

View full text Add to dashboard Cite

show abstract

“…The lexicon with hand-labeled syllable boundary information was provided by the Friedrich Alexander University (FAU). It is a result of merging lexica that were created for the research projects EVAR [14], Verbmobil [26], SmartKom [19] and SmartWeb [25]. Guidelines used in its creation closely match the ones described in [4].…”

Section: Datamentioning

confidence: 99%

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

Bayerl

Riedhammer

2019

Text, Speech, and Dialogue

View full text Add to dashboard Cite

This paper presents a comparison of a traditional hybrid speech recognition system (kaldi using WFST and TDNN with lattice-free MMI) and a lexiconfree 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 4gram 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-ofvocabulary techniques, keyword search indexing and medical speech processing.

show abstract