Blakus's picture
Update README.md
c781a07 verified
metadata
license: cc-by-sa-4.0
task_categories:
  - text-to-speech
language:
  - es
pretty_name: Latinoamerican Spanish Voice Dataset 🎙️

Compiled, and curated from the Crowdsourced high-quality speech datasets made by Google, available at https://openslr.org/resources.php.

This dataset consists of a wavs folder with the audios plus a .txt file with the path to the audio and the speaker's transcription.

Example:

wavs/vem_05223_00896110924.wav|Los corazones de pollo son una delicia.

wavs/vem_04310_01196944169.wav|Es un plato muy nutritivo.

wavs/vem_02484_00854567505.wav|En este momento estoy enviando a sus mails unos links de unas meditaciones en YouTube.

The audios are perfectly trimmed without any silence, and they are high quality: 48 Khz - 16 bits.

The names of the audios indicate the nationality of the speaker and whether the speaker is male or female.

There are at least 2 or 3 speakers of each sex for each nationality, with varying accents and voice timbres.

The nationalities are:

Argentina (ar)

Chile (cl)

Peru (per)

Colombia (co)

Venezuela (ve)

Puerto Rico (pr)

In total, there are approximately 24 hours of speech.