In this paper, we propose DASpeech, a non-autoregressive direct S2ST model which realizes both fast and high-quality S2ST.

Much of the promise of these models relies on learned concepts being high quality .

Contrastive learning has recently emerged as a promising approach for learning data representations that discover and disentangle the explanatory factors of the data.

Our primary focus is the identification of the latent structure in measurement models without parametric assumptions such as linearity or Gaussianity.

However, there still exist many properties of contrastive learning that are not guaranteed.