metadata

task_ids:
  - tts
language:
  - it
task_categories:
  - tts
multilinguality:
  - monolingual

Italian Male Voice

This dataset is an Italian version of LJSpeech, that merge all female audio of the same speaker finded into M-AILABS Speech Dataset.

This dataset contains 8h 23m of one speacker recorded at 16000Hz. This is a valid choiche to train an italian TTS deep model with female voice.