paper-with-me

홈 › Papers

Super Monotonic Alignment Search

2024-09-12 · Junhyeok Lee, Hyeongju Kim

Monotonic alignment search (MAS), introduced by Glow-TTS, is one of the most popular algorithm in TTS 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 paths, the time complexity of the algorithm is $O(T \times S)$. 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 \url{https://github.com/supertone-inc/super-monotonic-align}.

📄 PDF Abstract BibTeX arXiv:2409.07704

Code (1)

supertone-inc/super-monotonic-align 공식 구현 pytorch

Tasks

CPUGPU

Methods 이 논문이 사용한 방법론

Affine Coupling 설명 없음
Normalizing Flows Normalizing Flows are a method for constructing complex distributions by transforming a probability density through a series of invertible mappings. By repeatedly applying…
Invertible 1x1 Convolution The Invertible 1x1 Convolution is a type of convolution used in flow-based generative models that reverses the ordering of…
Activation Normalization Activation Normalization is a type of normalization used for flow-based generative models; specifically it was introduced in the GLOW…
GLOW 설명 없음
Glow-TTS Glow-TTS is a flow-based generative model for parallel TTS that does not require any external aligner. By combining the properties of flows and dynamic programming, the…
MAS This optimizer mix ADAM and SGD creating the MAS optimizer.

Similar Papers 제목 키워드 기반

Improvement in Sign Language Translation Using Text CTC Alignment

2024-12-12 · Sihan Tan, Taro Miyazaki, Nabeela Khan, Kazuhiro Nakadai

Current sign language translation (SLT) approaches often rely on gloss-based supervision with Connectionist Temporal Classification (CTC), limiting their ability to handle non-monotonic alignments between sign language v…

Sign Language TranslationTransfer LearningTranslation

Glow-TTS: A Generative Flow for Text-to-Speech via Monotonic Alignment Search

2020-05-22 · NeurIPS 2020 12 · Jaehyeon Kim, Sungwon Kim, Jungil Kong, Sungroh Yoon

Recently, text-to-speech (TTS) models such as FastSpeech and ParaNet have been proposed to generate mel-spectrograms from text in parallel. Despite the advantages, the parallel TTS models cannot be trained without guidan…

text-to-speechText to SpeechText-To-Speech Synthesis

Non-Monotonic Sentence Alignment via Semisupervised Learning

2013-08-01 · ACL 2013 8 · Xiaojun Quan, Chunyu Kit, Yan Song
Information RetrievalMachine TranslationSentence

Learning Monotonic Attention in Transducer for Streaming Generation

2024-11-26 · Zhengrui Ma, Yang Feng, Min Zhang

Streaming generation models are increasingly utilized across various fields, with the Transducer architecture being particularly popular in industrial applications. However, its input-synchronous decoding mechanism prese…

Non-Monotonic Latent Alignments for CTC-Based Non-Autoregressive Machine Translation

2022-10-08 · Chenze Shao, Yang Feng

Non-autoregressive translation (NAT) models are typically trained with the cross-entropy loss, which forces the model outputs to be aligned verbatim with the target sentence and will highly penalize small shifts in word …

Machine TranslationSentenceTranslation