FullStop:Punctuation and Segmentation Prediction for Dutch with Transformers

TL;DR

This paper introduces a Transformer-based Dutch punctuation and segmentation prediction system for unsegmented ASR outputs, significantly improving over baseline methods and making speech recognition results more readable and usable.

Contribution

It presents a new Dutch punctuation and segmentation model based on RobBERT, extending multilingual models, and demonstrates superior performance over baseline approaches.

Findings

Model outperforms machine translation baseline

Effective on out-of-domain data with sliding window approach

Publicly available Dutch punctuation prediction system

Abstract

When applying automated speech recognition (ASR) for Belgian Dutch (Van Dyck et al. 2021), the output consists of an unsegmented stream of words, without any punctuation. A next step is to perform segmentation and insert punctuation, making the ASR output more readable and easy to manually correct. As far as we know there is no publicly available punctuation insertion system for Dutch that functions at a usable level. The model we present here is an extension of the models of Guhr et al. (2021) for Dutch and is made publicly available. We trained a sequence classification model, based on the Dutch language model RobBERT (Delobelle et al. 2020). For every word in the input sequence, the models predicts a punctuation marker that follows the word. We have also extended a multilingual model, for cases where the language is unknown or where code switching applies. When performing the task of segmentation, the application of the best models onto out of domain test data, a sliding window of 200 words of the ASR output stream is sent to the classifier, and segmentation is applied when the system predicts a segmenting punctuation sign with a ratio above threshold. Results show to be much better than a machine translation baseline approach.