SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing

TL;DR

SentencePiece introduces a language-independent subword tokenizer and detokenizer that can be trained directly from raw text, enabling end-to-end neural text processing with comparable translation accuracy.

Contribution

It provides a novel, end-to-end subword segmentation tool that works directly on raw sentences, unlike previous tools requiring pre-tokenized input.

Findings

Achieves comparable accuracy to traditional methods in English-Japanese NMT

Supports training directly from raw sentences, simplifying preprocessing

Available as open-source software for easy adoption

Abstract

This paper describes SentencePiece, a language-independent subword tokenizer and detokenizer designed for Neural-based text processing, including Neural Machine Translation. It provides open-source C++ and Python implementations for subword units. While existing subword segmentation tools assume that the input is pre-tokenized into word sequences, SentencePiece can train subword models directly from raw sentences, which allows us to make a purely end-to-end and language independent system. We perform a validation experiment of NMT on English-Japanese machine translation, and find that it is possible to achieve comparable accuracy to direct subword training from raw sentences. We also compare the performance of subword training and segmentation with various configurations. SentencePiece is available under the Apache 2 license at https://github.com/google/sentencepiece.