paper-with-me

Papers

A GPU-based WFST Decoder with Exact Lattice Generation

2018-04-09 · Zhehuai Chen, Justin Luitjens, Hainan Xu, Yiming Wang, Daniel Povey, Sanjeev Khudanpur

We describe initial work on an extension of the Kaldi toolkit that supports weighted finite-state transducer (WFST) decoding on Graphics Processing Units (GPUs). We implement token recombination as an atomic GPU operation in order to fully parallelize the Viterbi beam search, and propose a dynamic load balancing strategy for more efficient token passing scheduling among GPU threads. We also redesign the exact lattice generation and lattice pruning algorithms for better utilization of the GPUs. Experiments on the Switchboard corpus show that the proposed method achieves identical 1-best results and lattice quality in recognition and confidence measure tasks, while running 3 to 15 times faster than the single process Kaldi decoder. The above results are reported on different GPU architectures. Additionally we obtain a 46-fold speedup with sequence parallelism and multi-process service (MPS) in GPU.

📄 PDF Abstract BibTeX arXiv:1804.03243

Code (0)

등록된 구현이 없습니다.

Tasks

DecoderGPUScheduling

Methods 이 논문이 사용한 방법론

Pruning 설명 없음

Similar Papers 제목 키워드 기반

GPU-Accelerated Viterbi Exact Lattice Decoder for Batched Online and Offline Speech Recognition

2019-10-22 · Hugo Braun, Justin Luitjens, Ryan Leary, Tim Kaldewey 외

We present an optimized weighted finite-state transducer (WFST) decoder capable of online streaming and offline batch processing of audio using Graphics Processing Units (GPUs). The decoder is efficient in memory utiliza…

CPUDecoderGPUspeech-recognition+1

Integration of TensorFlow based Acoustic Model with Kaldi WFST Decoder

2019-06-21 · Minkyu Lim, Ji-Hwan Kim

While the Kaldi framework provides state-of-the-art components for speech recognition like feature extraction, deep neural network (DNN)-based acoustic models, and a weighted finite state transducer (WFST)-based decoder,…

Decoderspeech-recognitionSpeech Recognition

WNARS: WFST based Non-autoregressive Streaming End-to-End Speech Recognition

2021-04-08 · Zhichao Wang, Wenwen Yang, Pan Zhou, Wei Chen

Recently, attention-based encoder-decoder (AED) end-to-end (E2E) models have drawn more and more attention in the field of automatic speech recognition (ASR). AED models, however, still have drawbacks when deploying in c…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Decoderspeech-recognition+1

Powerful and Extensible WFST Framework for RNN-Transducer Losses

2023-03-18 · Aleksandr Laptev, Vladimir Bataev, Igor Gitman, Boris Ginsburg

This paper presents a framework based on Weighted Finite-State Transducers (WFST) to simplify the development of modifications for RNN-Transducer (RNN-T) loss. Existing implementations of RNN-T use CUDA-related code, whi…

Sequence-to-Lattice Models for Fast Translation

2021-11-01 · Findings (EMNLP) 2021 11 · Yuntian Deng, Alexander Rush

Non-autoregressive machine translation (NAT) approaches enable fast generation by utilizing parallelizable generative processes. The remaining bottleneck in these models is their decoder layers; unfortunately unlike in a…

DecoderMachine TranslationTranslation