Self-supervised language learning from raw audio: Lessons from the Zero Resource Speech Challenge

Abstract

Recent progress in self-supervised or unsupervised machine learning has opened the possibility of building a full speech processing system from raw audio without using any textual representations or expert labels such as phonemes, dictionaries or parse trees. The contribution of the Zero Resource Speech Challenge series since 2015 has been to break down this long-term objective into four well-defined tasks -- Acoustic Unit Discovery, Spoken Term Discovery, Discrete Resynthesis, and Spoken Language Modeling -- and introduce associated metrics and benchmarks enabling model comparison and cumulative progress. We present an overview of the six editions of this challenge series since 2015, discuss the lessons learned, and outline the areas which need more work or give puzzling results.

References in corpus (7)

• Data Augmenting Contrastive Learning of Speech Representations in the Time Domain
• The Zero Resource Speech Benchmark 2021: Metrics and baselines for unsupervised spoken language modeling
• Self-Supervised Representation Learning for Speech Using Visual Grounding and Masked Language Modeling
• Predicting non-native speech perception using the Perceptual Assimilation Model and state-of-the-art acoustic models
• Zero resource speech synthesis using transcripts derived from perceptual acoustic units
• A Brief Overview of Unsupervised Neural Speech Representation Learning
• A Hierarchical Subspace Model for Language-Attuned Acoustic Unit Discovery