Model Card for Model ID

This BERT model classifies a dialogue system's user utterance as fluent or disfluent.

Model Details

Model Description

Developed by: 4i Intelligent Insights
Model type: BERT base cased
Language(s) (NLP): English
License: cc-by-nc-sa-4.0

Model Sources

Repository: http://research.4i.ai/code/BERT_disfluency_cls
Paper: https://aclanthology.org/2023.findings-acl.728/

Uses

The model is intended to be used for classifying English utterances of users interacting with a dialogue system. In our evaluation, the user utterances were speech transcriptions.

Out-of-Scope Use

This model has not been evaluated to be used on machine-generated text.

Bias, Risks, and Limitations

This model may not be accurate with non-native English speakers.

Training Data

The model has been fine-tuned on the Fisher English Corpus: http://github.com/joshua-decoder/fisher-callhome-corpus