论文标题
使用增强学习的神经序列到序列模型的语音递增文本
Incremental Text to Speech for Neural Sequence-to-Sequence Models using Reinforcement Learning
论文作者
论文摘要
现代文本到语音的方法需要在合成任何音频之前对整个输入字符序列进行处理。这种延迟限制了此类模型对时间敏感任务（例如同时解释）的适用性。将阅读角色与合成音频的作用相结合，从而降低了这一延迟。但是，这种交错动作序列的顺序在句子之间各不相同，这提出了应如何选择动作的问题。我们提出了一个基于加强学习的框架，以培训代理做出这一决定。我们将我们的绩效与确定性，基于规则的系统的绩效进行比较。我们的结果表明，我们的代理商成功地平衡了音频产生的延迟与合成音频的质量之间的权衡。更广泛地说，我们表明神经序列到序列模型可以适应以增量的方式运行。
Modern approaches to text to speech require the entire input character sequence to be processed before any audio is synthesised. This latency limits the suitability of such models for time-sensitive tasks like simultaneous interpretation. Interleaving the action of reading a character with that of synthesising audio reduces this latency. However, the order of this sequence of interleaved actions varies across sentences, which raises the question of how the actions should be chosen. We propose a reinforcement learning based framework to train an agent to make this decision. We compare our performance against that of deterministic, rule-based systems. Our results demonstrate that our agent successfully balances the trade-off between the latency of audio generation and the quality of synthesised audio. More broadly, we show that neural sequence-to-sequence models can be adapted to run in an incremental manner.
