Universal Automatic Phonetic Transcription into the International Phonetic Alphabet

TL;DR

This paper introduces a high-quality, language-agnostic speech-to-IPA transcription model based on wav2vec 2.0, capable of automating phonetic transcription with accuracy close to human annotators, aiding language documentation.

Contribution

The paper presents a universal speech-to-IPA model trained on high-quality data from multiple languages, achieving comparable or superior results to previous models despite using less data.

Findings

Model achieves near-human transcription quality.

Uses smaller, higher-quality training data.

Performs well across multiple languages.

Abstract

This paper presents a state-of-the-art model for transcribing speech in any language into the International Phonetic Alphabet (IPA). Transcription of spoken languages into IPA is an essential yet time-consuming process in language documentation, and even partially automating this process has the potential to drastically speed up the documentation of endangered languages. Like the previous best speech-to-IPA model (Wav2Vec2Phoneme), our model is based on wav2vec 2.0 and is fine-tuned to predict IPA from audio input. We use training data from seven languages from CommonVoice 11.0, transcribed into IPA semi-automatically. Although this training dataset is much smaller than Wav2Vec2Phoneme's, its higher quality lets our model achieve comparable or better results. Furthermore, we show that the quality of our universal speech-to-IPA models is close to that of human annotators.