Sequence-Level Unsupervised Training in Speech Recognition: A Theoretical Study

arXiv (Cornell University)

Abstract

Unsupervised speech recognition is a task of training a speech recognition model with unpaired data. To determine when and how unsupervised speech recognition can succeed, and how classification error relates to candidate training objectives, we develop a theoretical framework for unsupervised speech recognition grounded in classification error bounds. We introduce two conditions under which unsupervised speech recognition is possible. The necessity of these conditions are also discussed. Under these conditions, we derive a classification error bound for unsupervised speech recognition and validate this bound in simulations. Motivated by this bound, we propose a single-stage sequence-level cross-entropy loss for unsupervised speech recognition.

Authors 4

1. Zijian Yang Aachen Human Language Technology and Pattern Recognition Computer Science Department

   RWTH Aachen University

   Affiliation as printed

   Human Language Technology and Pattern Recognition , Computer Science Department RWTH Aachen University , Germany

2. Jörg Barkoczi Aachen Human Language Technology and Pattern Recognition Computer Science Department

   RWTH Aachen University

   Affiliation as printed

   Human Language Technology and Pattern Recognition , Computer Science Department RWTH Aachen University , Germany

3. Ralf Schlüter Aachen Human Language Technology and Pattern Recognition Computer Science Department

   RWTH Aachen University

   Affiliation as printed

   AppTek GmbH , Germany

   Human Language Technology and Pattern Recognition , Computer Science Department RWTH Aachen University , Germany

4. HERMANN NEY Aachen Human Language Technology and Pattern Recognition Computer Science Department

   RWTH Aachen University

   Affiliation as printed

   AppTek GmbH , Germany

   Human Language Technology and Pattern Recognition , Computer Science Department RWTH Aachen University , Germany