Bitte benutzen Sie diese Kennung, um auf die Ressource zu verweisen:
https://doi.org/10.21256/zhaw-30250
Langanzeige der Metadaten
DC Element | Wert | Sprache |
---|---|---|
dc.contributor.author | Bollinger, Tobias | - |
dc.contributor.author | Deriu, Jan Milan | - |
dc.contributor.author | Vogel, Manfred | - |
dc.date.accessioned | 2024-03-15T15:52:14Z | - |
dc.date.available | 2024-03-15T15:52:14Z | - |
dc.date.issued | 2023-06 | - |
dc.identifier.uri | https://digitalcollection.zhaw.ch/handle/11475/30250 | - |
dc.description.abstract | In this work, we studied the synthesis of Swiss German speech using different Text-to-Speech (TTS) models. We evaluated the TTS models on three corpora, and we found, that VITS models performed best, hence, using them for further testing. We also introduce a new method to evaluate TTS models by letting the discriminator of a trained vocoder GAN model predict whether a given waveform is human or synthesized. In summary, our best model delivers speech synthesis for different Swiss German dialects with previously unachieved quality. | de_CH |
dc.language.iso | en | de_CH |
dc.publisher | arXiv | de_CH |
dc.rights | http://creativecommons.org/licenses/by/4.0/ | de_CH |
dc.subject | Speech synthesis | de_CH |
dc.subject | Text to speech | de_CH |
dc.subject.ddc | 410.285: Computerlinguistik | de_CH |
dc.subject.ddc | 430: Deutsch | de_CH |
dc.title | Text-to-speech pipeline for Swiss German : a comparison | de_CH |
dc.type | Konferenz: Paper | de_CH |
dcterms.type | Text | de_CH |
zhaw.departement | School of Engineering | de_CH |
zhaw.organisationalunit | Centre for Artificial Intelligence (CAI) | de_CH |
dc.identifier.doi | 10.48550/arXiv.2305.19750 | de_CH |
dc.identifier.doi | 10.21256/zhaw-30250 | - |
zhaw.conference.details | 8th Swiss Text Analytics Conference – SwissText 2023, Neuchâtel, Switzerland, 12-14 June 2023 | de_CH |
zhaw.funding.eu | No | de_CH |
zhaw.originated.zhaw | Yes | de_CH |
zhaw.publication.status | publishedVersion | de_CH |
zhaw.publication.review | Peer review (Publikation) | de_CH |
zhaw.funding.snf | 200729 | de_CH |
zhaw.webfeed | Natural Language Processing | de_CH |
zhaw.funding.zhaw | End-to-End Low-Resource Speech Translation for Swiss German Dialects | de_CH |
zhaw.author.additional | No | de_CH |
zhaw.display.portrait | Yes | de_CH |
Enthalten in den Sammlungen: | Publikationen School of Engineering |
Dateien zu dieser Ressource:
Datei | Beschreibung | Größe | Format | |
---|---|---|---|---|
2023_Bollinger-etal_Text-to-speech-pipeline-for-Swiss-German.pdf | 1.03 MB | Adobe PDF | Öffnen/Anzeigen |
Zur Kurzanzeige
Bollinger, T., Deriu, J. M., & Vogel, M. (2023, June). Text-to-speech pipeline for Swiss German : a comparison. 8th Swiss Text Analytics Conference – SwissText 2023, Neuchâtel, Switzerland, 12-14 June 2023. https://doi.org/10.48550/arXiv.2305.19750
Bollinger, T., Deriu, J.M. and Vogel, M. (2023) ‘Text-to-speech pipeline for Swiss German : a comparison’, in 8th Swiss Text Analytics Conference – SwissText 2023, Neuchâtel, Switzerland, 12-14 June 2023. arXiv. Available at: https://doi.org/10.48550/arXiv.2305.19750.
T. Bollinger, J. M. Deriu, and M. Vogel, “Text-to-speech pipeline for Swiss German : a comparison,” in 8th Swiss Text Analytics Conference – SwissText 2023, Neuchâtel, Switzerland, 12-14 June 2023, Jun. 2023. doi: 10.48550/arXiv.2305.19750.
BOLLINGER, Tobias, Jan Milan DERIU und Manfred VOGEL, 2023. Text-to-speech pipeline for Swiss German : a comparison. In: 8th Swiss Text Analytics Conference – SwissText 2023, Neuchâtel, Switzerland, 12-14 June 2023. Conference paper. arXiv. Juni 2023
Bollinger, Tobias, Jan Milan Deriu, and Manfred Vogel. 2023. “Text-to-Speech Pipeline for Swiss German : A Comparison.” Conference paper. In 8th Swiss Text Analytics Conference – SwissText 2023, Neuchâtel, Switzerland, 12-14 June 2023. arXiv. https://doi.org/10.48550/arXiv.2305.19750.
Bollinger, Tobias, et al. “Text-to-Speech Pipeline for Swiss German : A Comparison.” 8th Swiss Text Analytics Conference – SwissText 2023, Neuchâtel, Switzerland, 12-14 June 2023, arXiv, 2023, https://doi.org/10.48550/arXiv.2305.19750.
Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt, soweit nicht anderweitig angezeigt.