SylReg

Dataset for SylReg This repository contains the datasets and alignments associated with the paper Speaker-Disentangled Chunk-Wise Regression for Syllabic Tokenization. Dataset Details Language: English License: CC BY-NC-SA 4.0 Repository: GitHub Code Paper: arXiv:2607.04064 Demo / Project Page: Project Page Sample Usage Below is an example of syllabic tokenization for speech language models using the trained models from the corresponding code… See the full description on the dataset page:

Source: Hugging Face Hub (ryota-komatsu/SylReg). Metadata imported from the dataset’s Hub tags.