基于自回归Transformer的文本到语音中的鲁棒性与无界长度泛化
Robust and Unbounded Length Generalization in Autoregressive Transformer-Based Text-to-Speech
AI总结：
本文提出Very Attentive Tacotron，通过可学习的对齐机制为交叉注意力提供相对位置信息，解决了自回归Transformer TTS在长语音上的重复、丢词和长度泛化问题，同时保持与T5基线相当的自然度。
摘要：
众所周知，基于自回归（AR）Transformer的序列模型难以泛化到比训练期间所见更长的序列。当应用于文本到语音（TTS）时，这些模型倾向于丢失或重复单词，或产生不稳定的输出，尤其是对于较长的语音。在本文中，我们引入了针对基于AR Transformer的编码器-解码器TTS系统的增强方法，以解决这些鲁棒性和长度泛化问题。我们的方法使用一种对齐机制，为交叉注意力操作提供相对位置信息。相关的对齐位置通过反向传播作为模型的潜在属性进行学习，并且在训练期间不需要外部对齐信息。虽然该方法针对TTS输入-输出对齐的单调性质进行了定制，但它仍然能够受益于交错的多头自注意力和交叉注意力操作的灵活建模能力。一个包含这些改进的系统，我们称之为Very Attentive Tacotron，在自然度和表现力上与基于T5的基线TTS系统相当，同时消除了重复或丢失单词的问题，并能够泛化到任意实际的语音长度。
英文摘要：
Autoregressive (AR) Transformer-based sequence models are known to have difficulty generalizing to sequences longer than those seen during training. When applied to text-to-speech (TTS), these models tend to drop or repeat words or produce erratic output, especially for longer utterances. In this paper, we introduce enhancements aimed at AR Transformer-based encoder-decoder TTS systems that address these robustness and length generalization issues. Our approach uses an alignment mechanism to provide cross-attention operations with relative location information. The associated alignment position is learned as a latent property of the model via backpropagation and requires no external alignment information during training. While the approach is tailored to the monotonic nature of TTS input-output alignment, it is still able to benefit from the flexible modeling power of interleaved multi-head self- and cross-attention operations. A system incorporating these improvements, which we call Very Attentive Tacotron, matches the naturalness and expressiveness of a baseline T5-based TTS system, while eliminating problems with repeated or dropped words and enabling generalization to any practical utterance length.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
