Towards Unsupervised Speech Recognition Without Pronunciation Models

TL;DR

This paper introduces a novel unsupervised speech recognition method that does not rely on pronunciation models or paired data, achieving competitive accuracy through joint speech and text modeling.

Contribution

It proposes a new approach for word-level unsupervised ASR that removes the need for phoneme lexicons and demonstrates its effectiveness on English speech data.

Findings

Achieves 20-23% word error rate without parallel transcripts

Outperforms previous lexicon-free unsupervised ASR models

Successfully refines word segmentation iteratively

Abstract

Recent advancements in supervised automatic speech recognition (ASR) have achieved remarkable performance, largely due to the growing availability of large transcribed speech corpora. However, most languages lack sufficient paired speech and text data to effectively train these systems. In this article, we tackle the challenge of developing ASR systems without paired speech and text corpora by proposing the removal of reliance on a phoneme lexicon. We explore a new research direction: word-level unsupervised ASR, and experimentally demonstrate that an unsupervised speech recognizer can emerge from joint speech-to-speech and text-to-text masked token-infilling. Using a curated speech corpus containing a fixed number of English words, our system iteratively refines the word segmentation structure and achieves a word error rate of between 20-23%, depending on the vocabulary size, without parallel transcripts, oracle word boundaries, or a pronunciation lexicon. This innovative model surpasses the performance of previous unsupervised ASR models under the lexicon-free setting.