Work

A Theory of Unsupervised Speech Recognition

Year: 2023

Type: conference-paper

Abstract: Unsupervised speech recognition (ASR-U) is the problem of learning automatic speech recognition (ASR) systems from unpaired speech-only and text-only corpora.While various algorithms exist to solve this problem, a theoretical framework is missing to study their properties and address such issues as sensitivity to hyperparameters and training instability.In this paper, we proposed a general theoretical framework to study the properties of ASR-U systems based on random matrix theory and the theory of neural tangent kernels.Such a framework allows us to prove various learnability conditions and sample complexity bounds of ASR-U.Extensive ASR-U experiments on synthetic languages with three classes of transition graphs provide strong empirical evidence for our theory (code available at cactuswith- thoughts/UnsupASRTheory.git).

Source: Annual Meeting of the Association for Computational Linguistics (ACL)

DOI: 10.18653/v1/2023.acl-long.67

Authors: Liming Wang, Mark Hasegawa-Johnson, Chang Yoo

Institutions: University of Illinois Urbana-Champaign, Korea Advanced Institute of Science and Technology

Language: English

Topic: Speech Recognition and Synthesis

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funders: Ministry of Science and ICT, South Korea, Institute for Information and Communications Technology Promotion

Awards: G3450174791, G4700831490, G4812080362, G6758007335