The Zero Resource Speech Challenge 2019: TTS without T. Dunbar, E., Algayres, R., Karadayi, J., Bernard, M., Benjumea, J., Cao, X., Miskic, L., Dugrain, C., Ondel, L., Black, A. W., Besacier, L., Sakti, S., & Dupoux, E. In September, 2019. ![The Zero Resource Speech Challenge 2019: TTS without T [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex We present the Zero Resource Speech Challenge 2019, which proposes to build a speech synthesizer without any text or pho-netic labels: hence, TTS without T (text-to-speech without text). We provide raw audio for a target voice in an unknown language (the Voice dataset), but no alignment, text or labels. Participants must discover subword units in an unsupervised way (using the Unit Discovery dataset) and align them to the voice recordings in a way that works best for the purpose of synthesizing novel utterances from novel speakers, similar to the target speaker's voice. We describe the metrics used for evaluation , a baseline system consisting of unsupervised subword unit discovery plus a standard TTS system, and a topline TTS using gold phoneme transcriptions. We present an overview of the 19 submitted systems from 10 teams and discuss the main results.
@inproceedings{dunbar_zero_2019,
title = {The {Zero} {Resource} {Speech} {Challenge} 2019: {TTS} without {T}},
shorttitle = {The {Zero} {Resource} {Speech} {Challenge} 2019},
url = {https://hal.science/hal-02274112},
abstract = {We present the Zero Resource Speech Challenge 2019, which proposes to build a speech synthesizer without any text or pho-netic labels: hence, TTS without T (text-to-speech without text). We provide raw audio for a target voice in an unknown language (the Voice dataset), but no alignment, text or labels. Participants must discover subword units in an unsupervised way (using the Unit Discovery dataset) and align them to the voice recordings in a way that works best for the purpose of synthesizing novel utterances from novel speakers, similar to the target speaker's voice. We describe the metrics used for evaluation , a baseline system consisting of unsupervised subword unit discovery plus a standard TTS system, and a topline TTS using gold phoneme transcriptions. We present an overview of the 19 submitted systems from 10 teams and discuss the main results.},
language = {en},
urldate = {2023-02-06},
author = {Dunbar, Ewan and Algayres, Robin and Karadayi, Julien and Bernard, Mathieu and Benjumea, Juan and Cao, Xuan-Nga and Miskic, Lucie and Dugrain, Charlotte and Ondel, Lucas and Black, Alan W. and Besacier, Laurent and Sakti, Sakriani and Dupoux, Emmanuel},
month = sep,
year = {2019},
}
Downloads: 0
{"_id":"38jYBFAtNv6evJwuk","bibbaseid":"dunbar-algayres-karadayi-bernard-benjumea-cao-miskic-dugrain-etal-thezeroresourcespeechchallenge2019ttswithoutt-2019","author_short":["Dunbar, E.","Algayres, R.","Karadayi, J.","Bernard, M.","Benjumea, J.","Cao, X.","Miskic, L.","Dugrain, C.","Ondel, L.","Black, A. W.","Besacier, L.","Sakti, S.","Dupoux, E."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"The Zero Resource Speech Challenge 2019: TTS without T","shorttitle":"The Zero Resource Speech Challenge 2019","url":"https://hal.science/hal-02274112","abstract":"We present the Zero Resource Speech Challenge 2019, which proposes to build a speech synthesizer without any text or pho-netic labels: hence, TTS without T (text-to-speech without text). We provide raw audio for a target voice in an unknown language (the Voice dataset), but no alignment, text or labels. Participants must discover subword units in an unsupervised way (using the Unit Discovery dataset) and align them to the voice recordings in a way that works best for the purpose of synthesizing novel utterances from novel speakers, similar to the target speaker's voice. We describe the metrics used for evaluation , a baseline system consisting of unsupervised subword unit discovery plus a standard TTS system, and a topline TTS using gold phoneme transcriptions. We present an overview of the 19 submitted systems from 10 teams and discuss the main results.","language":"en","urldate":"2023-02-06","author":[{"propositions":[],"lastnames":["Dunbar"],"firstnames":["Ewan"],"suffixes":[]},{"propositions":[],"lastnames":["Algayres"],"firstnames":["Robin"],"suffixes":[]},{"propositions":[],"lastnames":["Karadayi"],"firstnames":["Julien"],"suffixes":[]},{"propositions":[],"lastnames":["Bernard"],"firstnames":["Mathieu"],"suffixes":[]},{"propositions":[],"lastnames":["Benjumea"],"firstnames":["Juan"],"suffixes":[]},{"propositions":[],"lastnames":["Cao"],"firstnames":["Xuan-Nga"],"suffixes":[]},{"propositions":[],"lastnames":["Miskic"],"firstnames":["Lucie"],"suffixes":[]},{"propositions":[],"lastnames":["Dugrain"],"firstnames":["Charlotte"],"suffixes":[]},{"propositions":[],"lastnames":["Ondel"],"firstnames":["Lucas"],"suffixes":[]},{"propositions":[],"lastnames":["Black"],"firstnames":["Alan","W."],"suffixes":[]},{"propositions":[],"lastnames":["Besacier"],"firstnames":["Laurent"],"suffixes":[]},{"propositions":[],"lastnames":["Sakti"],"firstnames":["Sakriani"],"suffixes":[]},{"propositions":[],"lastnames":["Dupoux"],"firstnames":["Emmanuel"],"suffixes":[]}],"month":"September","year":"2019","bibtex":"@inproceedings{dunbar_zero_2019,\n\ttitle = {The {Zero} {Resource} {Speech} {Challenge} 2019: {TTS} without {T}},\n\tshorttitle = {The {Zero} {Resource} {Speech} {Challenge} 2019},\n\turl = {https://hal.science/hal-02274112},\n\tabstract = {We present the Zero Resource Speech Challenge 2019, which proposes to build a speech synthesizer without any text or pho-netic labels: hence, TTS without T (text-to-speech without text). We provide raw audio for a target voice in an unknown language (the Voice dataset), but no alignment, text or labels. Participants must discover subword units in an unsupervised way (using the Unit Discovery dataset) and align them to the voice recordings in a way that works best for the purpose of synthesizing novel utterances from novel speakers, similar to the target speaker's voice. We describe the metrics used for evaluation , a baseline system consisting of unsupervised subword unit discovery plus a standard TTS system, and a topline TTS using gold phoneme transcriptions. We present an overview of the 19 submitted systems from 10 teams and discuss the main results.},\n\tlanguage = {en},\n\turldate = {2023-02-06},\n\tauthor = {Dunbar, Ewan and Algayres, Robin and Karadayi, Julien and Bernard, Mathieu and Benjumea, Juan and Cao, Xuan-Nga and Miskic, Lucie and Dugrain, Charlotte and Ondel, Lucas and Black, Alan W. and Besacier, Laurent and Sakti, Sakriani and Dupoux, Emmanuel},\n\tmonth = sep,\n\tyear = {2019},\n}\n\n","author_short":["Dunbar, E.","Algayres, R.","Karadayi, J.","Bernard, M.","Benjumea, J.","Cao, X.","Miskic, L.","Dugrain, C.","Ondel, L.","Black, A. W.","Besacier, L.","Sakti, S.","Dupoux, E."],"key":"dunbar_zero_2019","id":"dunbar_zero_2019","bibbaseid":"dunbar-algayres-karadayi-bernard-benjumea-cao-miskic-dugrain-etal-thezeroresourcespeechchallenge2019ttswithoutt-2019","role":"author","urls":{"Paper":"https://hal.science/hal-02274112"},"metadata":{"authorlinks":{}},"html":""},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/mxmplx","dataSources":["aXmRAq63YsH7a3ufx"],"keywords":[],"search_terms":["zero","resource","speech","challenge","2019","tts","without","dunbar","algayres","karadayi","bernard","benjumea","cao","miskic","dugrain","ondel","black","besacier","sakti","dupoux"],"title":"The Zero Resource Speech Challenge 2019: TTS without T","year":2019}