Datasets:

proxectonos
/

CRPIH_UVigo-GL-Voices_extended

configs:
  - config_name: iago
    data_files:
      - split: all
        path: iago/iago_all.csv
      - split: train
        path: iago/iago_train.csv
      - split: test
        path: iago/iago_test.csv
    sep: "\t"
  - config_name: iago_extended
    data_files:
      - split: all
        path: iago_extended/iago_extended_all.csv
      - split: train
        path: iago_extended/iago_extended_train.csv
      - split: test
        path: iago_extended/iago_extended_test.csv
    sep: "\t"
  - config_name: icia
    data_files:
      - split: all
        path: icia/icia_all.csv
      - split: train
        path: icia/icia_train.csv
      - split: test
        path: icia/icia_test.csv
    sep: "\t"
  - config_name: paulo
    data_files:
      - split: all
        path: paulo/paulo_all.csv
      - split: train
        path: paulo/paulo_train.csv
      - split: test
        path: paulo/paulo_test.csv
    sep: "\t"
  - config_name: paulo_extended
    data_files:
      - split: all
        path: paulo_extended/paulo_extended_all.csv
      - split: train
        path: paulo_extended/paulo_extended_train.csv
      - split: test
        path: paulo_extended/paulo_extended_test.csv
    sep: "\t"
  - config_name: sabela
    data_files:
      - split: all
        path: sabela/sabela_all.csv
      - split: train
        path: sabela/sabela_train.csv
      - split: test
        path: sabela/sabela_test.csv
    sep: "\t"
license: cc-by-4.0
language:
  - gl
task_categories:
  - text-to-speech
tags:
  - male_speaker
  - female_speaker
pretty_name: CRPIH_UVigo-GL-Voices

CRPIH_UVigo-GL-Voices: Galician TTS dataset

CRPIH_UVigo-GL-Voices is a Galician TTS multi-speaker dataset containing audio recordings from four different speakers (two female and two male voices). The characteristics of each voice are detailed in the table below:

Voice name	Gender	Speaker	Recording	# Utts	Duration
Iago	Male	Amateur	Radio studio	1,316	1h 13min
Icía	Female	Amateur	Semi-professional studio	2,950	4h 5min
Paulo	Male	Amateur	Radio studio	1,316	1h 15min
Sabela	Female	Professional radio broadcaster	Professional studio	9,999	14h 28min

Sampling rate and audio format is 16kHz 16-bit PCM. A wider range of formats can be found in the CRPIH_UVigo-GL-Voices Zenodo corpus.

Each speaker recorded a subset of utterances from a text corpus of 10,000 sentences, with a length between 1 and 44 words. This corpus is mainly composed of press excerpts, but it also contains a small subset of manually designed sentences. The press excerpts were extracted from newspapers published before 2010 ("O Correo Galego", "Galicia Hoxe" and "Vieiros"), whereas the hand-crafted sentences were created at the CRPIH in the year 1999.

Corpus extension

For the smallest corpora, Iago and Paulo, an analysis has been developed to identify deficits in their quality. To address these shortcomings, these corpora have been expanded with new sentences, and the original ones that contained errors have been modified. All these audios were synthesised by the models Iago and Paulo.

Voice name	Gender	# Utts	Added	Modified	Duration
Iago	Male	1,667	351	87	1h 40min
Paulo	Male	1,666	350	44	1h 40min

Additional information

Acknowledgements

We would like to thank the speakers for recording and donating their voices.

Contact information

For further information, send an email to [email protected]

Licensing Information

Creative Commons Attribution 4.0 International

Funding

This research was produced within the framework of the Proxecto Nós, funded by the Ministry for Digital Transformation and Public Administration and the Recovery, Transformation, and Resilience Plan – Funded by the European Union – NextGenerationEU, as part of the Ilenia Project with reference 2022/TL22/00215336, and previously “The Nós project: Galician in the society and economy of Artificial Intelligence”, resulting from the agreement 2021-CP080 between the Xunta de Galicia and the University of Santiago de Compostela, and thanks to the Investigo program, within the National Recovery, Transformation and Resilience Plan, within the framework of the European Recovery Fund (NextGenerationEU).

Citation information

If you used the original CRPIH dataset, please cite as follows:

Centro Ramón Piñeiro para a Investigación en Humanidades (CRPIH), & Multimedia Technology Group (GTM) – atlanTTic Research Center for Telecommunication Technologies. (2023). CRPIH_UVigo-GL-Voices: Galician TTS dataset [Data set]. Zenodo. https://doi.org/10.5281/zenodo.8027725

If you used the extended dataset, please cite as follows:

Centro Ramón Piñeiro para a Investigación en Humanidades (CRPIH), Multimedia Technology Group (GTM) – atlanTTic Research Center for Telecommunication Technologies, María Pérez Lago, Noa Sobrino Lamas, Rebeca Rivas Gómez, Antonio Moscoso Sánchez, Carla Castedo, Adina Ioana Vladu, Carmen Magariños, Daniel Fernández López, Francisco Dubert-García, Xosé Luis Regueira Fernández & Elisa Fernández Rei. (2025). CRPIH_UVigo-GL-Voices_extended. [Data set]. HuggingFace. https://huggingface.co/datasets/proxectonos/CRPIH_UVigo-GL-Voices_extended/