Universal Automatic Phonetic Transcription into the International Phonetic Alphabet
This work addresses the time-consuming process of phonetic transcription for language documentation, particularly for endangered languages, though it is incremental as it builds on existing methods.
The paper tackles the problem of transcribing speech in any language into the International Phonetic Alphabet (IPA) by developing a state-of-the-art model based on wav2vec 2.0, achieving results comparable to or better than previous models and close to human annotators.
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.