Paper: arXiv
Authors: Yunchao He, Jian Luan, Yujun Wang
Abstract: Sequence expansion between encoder and decoder is a critical challenge in sequence-to-sequence TTS. Attention-based methods achieve great naturalness but suffer from unstable issues like missing and repeating phonemes, not to mention accurate duration control. Duration-informed methods, on the contrary, seem to easily adjust phoneme duration but show obvious degradation in speech naturalness. This paper proposes PAMA-TTS to address the problem. It takes the advantage of both flexible attention and explicit duration models. Based on the monotonic attention mechanism, PAMA-TTS also leverages token duration and relative position of a frame, especially countdown information, i.e. in how many future frames the present phoneme will end. They help the attention to move forward along the token sequence in a soft but reliable control. Experimental results prove that PAMA-TTS achieves the highest naturalness, while has on-par or even better duration controllability than the duration-informed model.
| 1: |
| 2: |
| 3: |
| 4: |
| 5: |
Random sample from test set.
| Proposed PAMA-TTS | Tacotron-length-regulator (TLR) | Tacotron-stepwise (TSW) |
|---|---|---|
| 1: 周霖先生您好，我这边是《小米贷款》审核机器人-小融，现在我们这边有一个为您提额的活动，需要您补充信息，请问您现在方便吗？ | ||
| 2: 好的，您目前正常经营的公司名称是下面哪一个，一、河北商玛商贸有限公司；二、汕尾市城区正力货运代理店；三、南京雄起服饰有限公司；四、以上都不是，请您选择一二三四任意一个 | ||
| 3: 请问您公司于哪一年成立？ | ||
| 4: 好的，您的主营业务是什么？ | ||
| 5: 好的，您本年度每个月平均销售额是下面哪一个，一、小于等于10万；二、11万（至）30万；三、31万（至）50万；四、50万以上，以上请您选择一二三四任意一个 | ||
| 6: 好的，您上个月的销售额是多少万元？ | ||
| 7: 抱歉，我没有听清，方便重复一下么？？？？？？？？？ | ||
| 8: 好的，您的毛利是下面哪一个？一、小于等于百分之5；二、百分之6（至）百分之8； 三、百分之9（至）百分之10；四、百分之10以上，以上请您选择一二三四任意一个 | ||
| 9: 好的，最后请您说下您的身份证号码后四位？ | ||
| 10: 好的非常感谢，本次已经核实身份完毕，祝您生活愉快，再见。 | ||
| Samples |
|---|
| Samples |
|---|