This dataset contains processed audio alignments from AAdonis/multilingual_audio_alignments (spanish).
This dataset uses mixed text/phoneme conditioning with a curriculum learning schedule:
p_start: 0.0 (starting probability of using phonemes)
p_end: 0.0 (ending probability of using phonemes)
curriculum_rows: 400000 (rows over which probability increases)
Early in the dataset, more words… See the full description on the dataset page:
https://huggingface.co/datasets/AdoCleanCode/SPEEED_s3_words_spanish_200k-400k.
