Comparative Layer-Wise Analysis of Self-Supervised Speech Models

Many self-supervised speech models, varying in their pre-training objective, input modality, and pre-training data, have been proposed in the last few years. Despite impressive successes on downstream tasks, we still have a limited understanding of the properties encoded by the models and the differ...

Full description

Saved in:
Bibliographic Details
Main Authors: Pasad, Ankita, Shi, Bowen, Livescu, Karen
Format: Conference Proceeding
Language:English
Subjects:
Analytical models
model analysis
Navigation
Representation learning
Self-supervised pre-training
Signal processing
Speech recognition
speech representation learning
Syntactics
Citations: Items that cite this one
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Many self-supervised speech models, varying in their pre-training objective, input modality, and pre-training data, have been proposed in the last few years. Despite impressive successes on downstream tasks, we still have a limited understanding of the properties encoded by the models and the differences across models. In this work, we examine the intermediate representations for a variety of recent models. Specifically, we measure acoustic, phonetic, and word-level properties encoded in individual layers, using a lightweight analysis tool based on canonical correlation analysis (CCA). We find that these properties evolve across layers differently depending on the model, and the variations relate to the choice of pre-training objective. We further investigate the utility of our analyses for downstream tasks by comparing the property trends with performance on speech recognition and spoken language understanding tasks. We discover that CCA trends provide reliable guidance to choose layers of interest for downstream tasks and that single-layer performance often matches or improves upon using all layers, suggesting implications for more efficient use of pre-trained models. 1
ISSN:2379-190X
DOI:10.1109/ICASSP49357.2023.10096149