RNN Approaches to Text Normalization: A Challenge

TL;DR

This paper challenges the effectiveness of RNNs for text normalization tasks, showing that while they perform well in accuracy, they produce problematic errors that can be mitigated with additional filtering, and introduces a new dataset for future research.

Contribution

It provides a new open-source dataset for text normalization and evaluates RNN architectures, highlighting limitations and proposing combined approaches for improved performance.

Findings

RNNs achieve high accuracy but generate critical errors.

A simple FST filter can significantly reduce RNN errors.

Pure RNN approaches may not suffice for reliable text normalization.

Abstract

This paper presents a challenge to the community: given a large corpus of written text aligned to its normalized spoken form, train an RNN to learn the correct normalization function. We present a data set of general text where the normalizations were generated using an existing text normalization component of a text-to-speech system. This data set will be released open-source in the near future. We also present our own experiments with this data set with a variety of different RNN architectures. While some of the architectures do in fact produce very good results when measured in terms of overall accuracy, the errors that are produced are problematic, since they would convey completely the wrong message if such a system were deployed in a speech application. On the other hand, we show that a simple FST-based filter can mitigate those errors, and achieve a level of accuracy not achievable by the RNN alone. Though our conclusions are largely negative on this point, we are actually not arguing that the text normalization problem is intractable using an pure RNN approach, merely that it is not going to be something that can be solved merely by having huge amounts of annotated text data and feeding that to a general RNN model. And when we open-source our data, we will be providing a novel data set for sequence-to-sequence modeling in the hopes that the the community can find better solutions. The data used in this work have been released and are available at: https://github.com/rwsproat/text-normalization-data