ASR-Paragraph-Breaks

ASR Paragraph Breaks Hand-annotated paragraph breaks for flat speech-to-text output. ASR engines emit punctuated, cased text as one block. This dataset pairs that block (text) with the same text split into paragraphs by a person (paragraphed). Nothing else changes: words, punctuation and casing are identical, and the annotation tool refuses any save that alters them. It is training data for a small text-in/text-out model that does only this one job. Paragraphing is a judgement… See the full description on the dataset page: https://huggingface.co/datasets/danielrosehill/ASR-Paragraph-Breaks.