facebook/s2t-medium-mustc-multilingual-st

The S2T model was proposed in this paper and released in this repository . It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are fed into the encoder . The model is trained with standard autoregressive cross-entropy loss and generates the translations autoregressively . This model can be used for end-to-end English speech to French text translation . The following example shows how to transate English to French and German text using the facebook/s2t-medium-mustc-multilingual-stcheckpoint. The model was trained on Mu-C-C is trained on a multilingual speech translation,