将语音合成中的对齐搜索算法加速72倍，支持长文本高效处理。
Super Monotonic Alignment Search
- 利用GPU并行化文本维度，避免跨设备拷贝开销
- 在极端长文本情况下速度提升最高达72倍
- 适合需要快速生成长语音的TTS系统开发者
Monotonic alignment search (MAS) 是文本到语音合成中估计文本与语音间未知对齐关系的主流算法。由于该算法需通过动态规划缓存所有可能路径来寻找最可能的对齐，时间复杂度为 $O(T \times S)$，其中 $T$ 为文本长度，$S$ 为语音表示长度。Glow-TTS作者在CPU上运行此算法，虽提及难以并行化，但研究发现其可在文本长度维度实现并行，且CPU执行时存在大量跨设备拷贝开销。为此，我们实现了Triton内核和PyTorch JIT脚本，使MAS在GPU上无跨设备拷贝地加速执行。结果表明，Super-MAS Triton内核在极端长文本情况下最快可达72倍加速。代码已开源：https://github.com/supertone-inc/super-monotonic-align。
原文摘要 · Abstract (English)
Monotonic alignment search (MAS), introduced by Glow-TTS, is one of the most popular algorithm in text-to-speech to estimate unknown alignments between text and speech. Since this algorithm needs to search for the most probable alignment with dynamic programming by caching all possible paths, the time complexity of the algorithm is $O(T \times S)$, where $T$ is the length of text and $S$ is the length of speech representation. The authors of Glow-TTS run this algorithm on CPU, and while they mentioned it is difficult to parallelize, we found that MAS can be parallelized in text length dimension and CPU execution consumes an inordinate amount of time for inter-device copy. Therefore, we implemented a Triton kernel and PyTorch JIT script to accelerate MAS on GPU without inter-device copy. As a result, Super-MAS Triton kernel is up to 72 times faster in the extreme-length case. The code is available at https://github.com/supertone-inc/super-monotonic-align.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。