Are discrete units necessary for Spoken Language Modeling?
This work addresses the problem of unsupervised spoken language modeling for researchers, showing that discrete representations are crucial for performance gains, though it is incremental as it builds on existing methods like HuBERT.
The study investigated whether discrete units are necessary for spoken language modeling, finding that discretization is essential as it removes linguistically irrelevant information and improves performance. The model trained on HuBERT discrete units achieved new state-of-the-art results in lexical, syntactic, and semantic metrics on the Zero Resource Speech Challenge 2021.
Recent work in spoken language modeling shows the possibility of learning a language unsupervisedly from raw audio without any text labels. The approach relies first on transforming the audio into a sequence of discrete units (or pseudo-text) and then training a language model directly on such pseudo-text. Is such a discrete bottleneck necessary, potentially introducing irreversible errors in the encoding of the speech signal, or could we learn a language model without discrete units at all? In this work, we study the role of discrete versus continuous representations in spoken language modeling. We show that discretization is indeed essential for good results in spoken language modeling. We show that discretization removes linguistically irrelevant information from the continuous features, helping to improve language modeling performances. On the basis of this study, we train a language model on the discrete units of the HuBERT features, reaching new state-of-the-art results in the lexical, syntactic and semantic metrics of the Zero Resource Speech Challenge 2021 (Track 1 - Speech Only).