End-to-End Automatic Speech Translation of Audiobooks
This work addresses speech translation for audiobook applications, but it is incremental as it builds on prior research with a modified training setup.
The paper tackles end-to-end speech-to-text translation for audiobooks, exploring a setup where source language transcription is available only during training, and shows that compact and efficient models can be trained in this configuration.
We investigate end-to-end speech-to-text translation on a corpus of audiobooks specifically augmented for this task. Previous works investigated the extreme case where source language transcription is not available during learning nor decoding, but we also study a midway case where source language transcription is available at training time only. In this case, a single model is trained to decode source speech into target text in a single pass. Experimental results show that it is possible to train compact and efficient end-to-end speech translation models in this setup. We also distribute the corpus and hope that our speech translation baseline on this corpus will be challenged in the future.