Papers by Flora Badin
Chunk Different Kind of Spoken Discourse: Challenges for Machine Learning (2020.lrec-1)
Copied to clipboard
| Challenge: | Existing chunkers for spoken data are based on a corpus composed of monologues and spontaneous talk in interaction. |
| Approach: | They propose to use CRFs to develop a chunker for spoken data . the chunker is based on a small corpus composed of two kinds of discourse . |
| Outcome: | The proposed chunker is based on a spoken corpus composed of monologue and spontaneous talk in interaction. |