Curriculum Pre-training for End-to-End Speech Translation
This addresses the problem of encoder burden in speech translation for researchers and practitioners, offering an incremental improvement over traditional pre-training methods.
The paper tackles the challenge of end-to-end speech translation by proposing a curriculum pre-training method that includes transcription, understanding, and cross-lingual mapping courses, resulting in significant improvements on En-De and En-Fr benchmarks.
End-to-end speech translation poses a heavy burden on the encoder, because it has to transcribe, understand, and learn cross-lingual semantics simultaneously. To obtain a powerful encoder, traditional methods pre-train it on ASR data to capture speech features. However, we argue that pre-training the encoder only through simple speech recognition is not enough and high-level linguistic knowledge should be considered. Inspired by this, we propose a curriculum pre-training method that includes an elementary course for transcription learning and two advanced courses for understanding the utterance and mapping words in two languages. The difficulty of these courses is gradually increasing. Experiments show that our curriculum pre-training method leads to significant improvements on En-De and En-Fr speech translation benchmarks.