Direct speech-to-speech translation (S2ST), in which all components can be optimized jointly, is advantageous over cascaded approaches to achieve fast inference with a simplified pipeline. We present a novel two-pass direct S2ST architecture, UnitY, which first generates textual representations and predicts discrete acoustic units subsequently. We enhance the model performance by subword prediction in the first-pass decoder, advanced two-pass decoder architecture design and search strategy, and better training regularization. To leverage large amounts of unlabeled text data, we pre-train the first-pass text decoder based on the self-supervised denoising auto-encoding task. Experimental evaluations on benchmark datasets at various data scales demonstrate that UnitY outperforms a single-pass speech-to-unit translation model by 2.5-4.2 ASR-BLEU with 2.83x decoding speed-up. We show that the proposed methods boost the performance even when predicting spectrogram in the second pass. However, predicting discrete units achieves 2.51x decoding speed-up compared to that case.

翻译：直接语音到语音翻译（S2ST）能够实现所有组件的联合优化，相比级联方法具有简化流程、实现快速推理的优势。我们提出了一种新颖的两遍式直接S2ST架构UnitY，该架构首先生成文本表征，随后预测离散声学单元。通过第一遍解码器中的子词预测、先进的两遍式解码器架构设计与搜索策略，以及更优的训练正则化方法，我们提升了模型性能。为利用海量无标注文本数据，我们基于自监督去噪自编码任务对第一遍文本解码器进行预训练。在不同数据规模的标准数据集上的实验评估表明，UnitY在解码速度提升2.83倍的情况下，相比单遍式语音到单元翻译模型实现了2.5-4.2的ASR-BLEU提升。我们证明，即使第二遍预测语谱图，所提出的方法仍能提升性能。然而，与预测语谱图的情况相比，预测离散单元实现了2.51倍的解码速度提升。