TaL: a synchronised multi-speaker corpus of ultrasound tongue imaging, audio, and lip videos

TL;DR

The TaL corpus is a publicly available multi-speaker dataset combining ultrasound tongue imaging, lip videos, and audio, designed for speech recognition, synthesis, and synchronisation research.

Contribution

This paper introduces the TaL corpus, a comprehensive multi-speaker dataset with synchronized ultrasound, video, and audio data for speech research.

Findings

Benchmark results for speech recognition on TaL corpus

Articulatory-to-acoustic mapping performance data

Automatic ultrasound-audio synchronisation results

Abstract

We present the Tongue and Lips corpus (TaL), a multi-speaker corpus of audio, ultrasound tongue imaging, and lip videos. TaL consists of two parts: TaL1 is a set of six recording sessions of one professional voice talent, a male native speaker of English; TaL80 is a set of recording sessions of 81 native speakers of English without voice talent experience. Overall, the corpus contains 24 hours of parallel ultrasound, video, and audio data, of which approximately 13.5 hours are speech. This paper describes the corpus and presents benchmark results for the tasks of speech recognition, speech synthesis (articulatory-to-acoustic mapping), and automatic synchronisation of ultrasound to audio. The TaL corpus is publicly available under the CC BY-NC 4.0 license.