Fetching the paper…

Unsupervised Speech Recognition via Segmental Empirical Output Distribution Matching · Around