English
Related papers

Related papers: Stateful Conformer with Cache-based Inference for …

200 papers

Attention-based models have been gaining popularity recently for their strong performance demonstrated in fields such as machine translation and automatic speech recognition. One major challenge of attention-based models is the need of…

Computation and Language · Computer Science 2020-11-17 Ching-Feng Yeh , Yongqiang Wang , Yangyang Shi , Chunyang Wu , Frank Zhang , Julian Chan , Michael L. Seltzer

A popular approach to streaming speech translation is to employ a single offline model with a wait-k policy to support different latency requirements, which is simpler than training multiple online models with different latency constraints.…

Computation and Language · Computer Science 2023-10-27 Biao Fu , Minpeng Liao , Kai Fan , Zhongqiang Huang , Boxing Chen , Yidong Chen , Xiaodong Shi

This paper investigates four types of cross-utterance speech contexts modeling approaches for streaming and non-streaming Conformer-Transformer (C-T) ASR systems: i) input audio feature concatenation; ii) cross-utterance Encoder embedding…

Audio and Speech Processing · Electrical Eng. & Systems 2025-08-15 Mingyu Cui , Mengzhe Geng , Jiajun Deng , Chengxi Deng , Jiawen Kang , Shujie Hu , Guinan Li , Tianzi Wang , Zhaoqing Li , Xie Chen , Xunying Liu

Streaming ASR with strict latency constraints is required in many speech recognition applications. In order to achieve the required latency, streaming ASR models sacrifice accuracy compared to non-streaming ASR models due to lack of future…

Computation and Language · Computer Science 2022-03-30 Jay Mahadeokar , Yangyang Shi , Ke Li , Duc Le , Jiedan Zhu , Vikas Chandra , Ozlem Kalinli , Michael L Seltzer

Recently, Transformer-based encoder-decoder models have demonstrated strong performance in multilingual speech recognition. However, the decoder's autoregressive nature and large size introduce significant bottlenecks during inference.…

Audio and Speech Processing · Electrical Eng. & Systems 2025-08-28 Yunkyu Lim , Jihwan Park , Hyung Yong Kim , Hanbin Lee , Byeong-Yeol Kim

Using end-to-end models for speech translation (ST) has increasingly been the focus of the ST community. These models condense the previously cascaded systems by directly converting sound waves into translated text. However, cascaded models…

Computation and Language · Computer Science 2021-01-25 Orion Weller , Matthias Sperber , Christian Gollan , Joris Kluivers

Transformer-based models have achieved state-of-the-art performance on speech translation tasks. However, the model architecture is not efficient enough for streaming scenarios since self-attention is computed over an entire input sequence…

Computation and Language · Computer Science 2020-11-03 Xutai Ma , Yongqiang Wang , Mohammad Javad Dousti , Philipp Koehn , Juan Pino

In interactive automatic speech recognition (ASR) systems, low-latency requirements limit the amount of search space that can be explored during decoding, particularly in end-to-end neural ASR. In this paper, we present a novel streaming…

Audio and Speech Processing · Electrical Eng. & Systems 2024-01-29 Denis Filimonov , Prabhat Pandey , Ariya Rastrow , Ankur Gandhe , Andreas Stolcke

The recurrent neural network transducer (RNN-T) is a prominent streaming end-to-end (E2E) ASR technology. In RNN-T, the acoustic encoder commonly consists of stacks of LSTMs. Very recently, as an alternative to LSTM layers, the Conformer…

We design an online end-to-end speech recognition system based on Time-Depth Separable (TDS) convolutions and Connectionist Temporal Classification (CTC). We improve the core TDS architecture in order to limit the future context and hence…

In this work, we present a hybrid CTC/Attention model based on a ResNet-18 and Convolution-augmented transformer (Conformer), that can be trained in an end-to-end manner. In particular, the audio and visual encoders learn to extract…

Computer Vision and Pattern Recognition · Computer Science 2021-02-15 Pingchuan Ma , Stavros Petridis , Maja Pantic

Models for streaming speech translation (ST) can achieve high accuracy and low latency if they're developed with vast amounts of paired audio in the source language and written text in the target language. Yet, these text labels for the…

Computation and Language · Computer Science 2024-10-08 Rui Zhao , Jinyu Li , Ruchao Fan , Matt Post

We propose to deliberate the hypothesis alignment of a streaming RNN-T model with the previously proposed Align-Refine non-autoregressive decoding method and its improved versions. The method performs a few refinement steps, where each step…

Computation and Language · Computer Science 2021-12-22 Weiran Wang , Ke Hu , Tara Sainath

We propose Chunk-wise Attention Transducer (CHAT), a novel extension to RNN-T models that processes audio in fixed-size chunks while employing cross-attention within each chunk. This hybrid approach maintains RNN-T's streaming capability…

Machine Learning · Computer Science 2026-03-02 Hainan Xu , Vladimir Bataev , Travis M. Bartley , Jagadeesh Balam

Encoder-decoder based sequence-to-sequence models have demonstrated state-of-the-art results in end-to-end automatic speech recognition (ASR). Recently, the transformer architecture, which uses self-attention to model temporal context…

Sound · Computer Science 2020-07-02 Niko Moritz , Takaaki Hori , Jonathan Le Roux

In this work we propose an inference technique, asynchronous revision, to unify streaming and non-streaming speech recognition models. Specifically, we achieve dynamic latency with only one model by using arbitrary right context during…

Audio and Speech Processing · Electrical Eng. & Systems 2020-11-04 Mingkun Huang , Meng Cai , Jun Zhang , Yang Zhang , Yongbin You , Yi He , Zejun Ma

Recently, end-to-end speech recognition with a hybrid model consisting of the connectionist temporal classification(CTC) and the attention encoder-decoder achieved state-of-the-art results. In this paper, we propose a novel CTC decoder…

Sound · Computer Science 2018-11-02 Zhe Yuan , Zhuoran Lyu , Jiwei Li , Xi Zhou

In this paper, we propose two novel approaches, which integrate long-content information into the factorized neural transducer (FNT) based architecture in both non-streaming (referred to as LongFNT ) and streaming (referred to as SLongFNT )…

Sound · Computer Science 2024-03-21 Xun Gong , Yu Wu , Jinyu Li , Shujie Liu , Rui Zhao , Xie Chen , Yanmin Qian

Sequence transducers, such as the RNN-T and the Conformer-T, are one of the most promising models of end-to-end speech recognition, especially in streaming scenarios where both latency and accuracy are important. Although various methods,…

Audio and Speech Processing · Electrical Eng. & Systems 2022-11-07 Yusuke Shinohara , Shinji Watanabe

Optimization of modern ASR architectures is among the highest priority tasks since it saves many computational resources for model training and inference. The work proposes a new Uconv-Conformer architecture based on the standard Conformer…

Audio and Speech Processing · Electrical Eng. & Systems 2023-03-14 Andrei Andrusenko , Rauf Nasretdinov , Aleksei Romanenko