@inproceedings{guevara-rukoz-etal-2020-crowdsourcing,
    title = "Crowdsourcing {L}atin {A}merican {S}panish for Low-Resource Text-to-Speech",
    author = "Guevara-Rukoz, Adriana  and
      Demirsahin, Isin  and
      He, Fei  and
      Chu, Shan-Hui Cathy  and
      Sarin, Supheakmungkol  and
      Pipatsrisawat, Knot  and
      Gutkin, Alexander  and
      Butryna, Alena  and
      Kjartansson, Oddur",
    editor = "Calzolari, Nicoletta  and
      B{\'e}chet, Fr{\'e}d{\'e}ric  and
      Blache, Philippe  and
      Choukri, Khalid  and
      Cieri, Christopher  and
      Declerck, Thierry  and
      Goggi, Sara  and
      Isahara, Hitoshi  and
      Maegaard, Bente  and
      Mariani, Joseph  and
      Mazo, H{\'e}l{\`e}ne  and
      Moreno, Asuncion  and
      Odijk, Jan  and
      Piperidis, Stelios",
    booktitle = "Proceedings of the Twelfth Language Resources and Evaluation Conference",
    month = may,
    year = "2020",
    address = "Marseille, France",
    publisher = "European Language Resources Association",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2020.lrec-1.801/",
    pages = "6504--6513",
    language = "eng",
    ISBN = "979-10-95546-34-4",
    abstract = "In this paper we present a multidialectal corpus approach for building a text-to-speech voice for a new dialect in a language with existing resources, focusing on various South American dialects of Spanish. We first present public speech datasets for Argentinian, Chilean, Colombian, Peruvian, Puerto Rican and Venezuelan Spanish specifically constructed with text-to-speech applications in mind using crowd-sourcing. We then compare the monodialectal voices built with minimal data to a multidialectal model built by pooling all the resources from all dialects. Our results show that the multidialectal model outperforms the monodialectal baseline models. We also experiment with a {\textquotedblleft}zero-resource{\textquotedblright} dialect scenario where we build a multidialectal voice for a dialect while holding out target dialect recordings from the training data."
}