no code implementations • 16 Jan 2024 • Jiyang Tang, Kwangyoun Kim, Suwon Shon, Felix Wu, Prashant Sridhar, Shinji Watanabe
Compared to studies with similar motivations, the proposed loss operates directly on the cross attention weights and is easier to implement.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +1
no code implementations • 15 Dec 2023 • Suwon Shon, Kwangyoun Kim, Prashant Sridhar, Yi-Te Hsu, Shinji Watanabe, Karen Livescu
Considering the recent advances in generative large language models (LLM), we hypothesize that an LLM could generate useful context information using the preceding text.
2 code implementations • 18 May 2023 • Yifan Peng, Kwangyoun Kim, Felix Wu, Brian Yan, Siddhant Arora, William Chen, Jiyang Tang, Suwon Shon, Prashant Sridhar, Shinji Watanabe
Conformer, a convolution-augmented Transformer variant, has become the de facto encoder architecture for speech processing due to its superior performance in various tasks, including automatic speech recognition (ASR), speech translation (ST) and spoken language understanding (SLU).
Automatic Speech Recognition Automatic Speech Recognition (ASR) +2
1 code implementation • 27 Feb 2023 • Yifan Peng, Kwangyoun Kim, Felix Wu, Prashant Sridhar, Shinji Watanabe
Self-supervised speech representation learning (SSL) has shown to be effective in various downstream tasks, but SSL models are usually large and slow.
no code implementations • 16 Dec 2022 • Suwon Shon, Felix Wu, Kwangyoun Kim, Prashant Sridhar, Karen Livescu, Shinji Watanabe
During the fine-tuning stage, we introduce an auxiliary loss that encourages this context embedding vector to be similar to context vectors of surrounding segments.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +5
1 code implementation • 30 Sep 2022 • Kwangyoun Kim, Felix Wu, Yifan Peng, Jing Pan, Prashant Sridhar, Kyu J. Han, Shinji Watanabe
Conformer, combining convolution and self-attention sequentially to capture both local and global information, has shown remarkable performance and is currently regarded as the state-of-the-art for automatic speech recognition (ASR).
Ranked #9 on Speech Recognition on LibriSpeech test-other
Automatic Speech Recognition Automatic Speech Recognition (ASR) +1
1 code implementation • 2 May 2022 • Felix Wu, Kwangyoun Kim, Shinji Watanabe, Kyu Han, Ryan Mcdonald, Kilian Q. Weinberger, Yoav Artzi
We introduce Wav2Seq, the first self-supervised approach to pre-train both parts of encoder-decoder models for speech data.
Ranked #3 on Named Entity Recognition (NER) on SLUE
Automatic Speech Recognition Automatic Speech Recognition (ASR) +6
no code implementations • 11 Oct 2021 • Jing Pan, Tao Lei, Kwangyoun Kim, Kyu Han, Shinji Watanabe
The Transformer architecture has been well adopted as a dominant architecture in most sequence transduction tasks including automatic speech recognition (ASR), since its attention mechanism excels in capturing long-range dependencies.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +4
1 code implementation • 14 Sep 2021 • Felix Wu, Kwangyoun Kim, Jing Pan, Kyu Han, Kilian Q. Weinberger, Yoav Artzi
This paper is a study of performance-efficiency trade-offs in pre-trained models for automatic speech recognition (ASR).
Automatic Speech Recognition Automatic Speech Recognition (ASR) +2
no code implementations • 17 Jun 2021 • Kwangyoun Kim, Felix Wu, Prashant Sridhar, Kyu J. Han, Shinji Watanabe
A Multi-mode ASR model can fulfill various latency requirements during inference -- when a larger latency becomes acceptable, the model can process longer future context to achieve higher accuracy and when a latency budget is not flexible, the model can be less dependent on future context but still achieve reliable accuracy.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +1
1 code implementation • 23 Jul 2020 • Kyungmin Lee, Hyunwhan Joe, Hyeontaek Lim, Kwangyoun Kim, Sungsoo Kim, Chang Woo Han, Hong-Gee Kim
Input sequences are capsulized then sliced by a window size.
no code implementations • 15 Feb 2020 • Chanwoo Kim, Kwangyoun Kim, Sathish Reddy Indurthi
More specifically, a time-frequency bin is masked if the filterbank energy in this bin is less than a certain energy threshold.
no code implementations • 2 Jan 2020 • Kwangyoun Kim, Kyungmin Lee, Dhananjaya Gowda, Junmo Park, Sungsoo Kim, Sichen Jin, Young-Yoon Lee, Jinsu Yeo, Daehyun Kim, Seokyeong Jung, Jungin Lee, Myoungji Han, Chanwoo Kim
In this paper, we present a new on-device automatic speech recognition (ASR) system based on monotonic chunk-wise attention (MoChA) models trained with large (> 10K hours) corpus.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +3
no code implementations • 28 Dec 2019 • Abhinav Garg, Dhananjaya Gowda, Ankur Kumar, Kwangyoun Kim, Mehul Kumar, Chanwoo Kim
In this paper, we propose a refined multi-stage multi-task training strategy to improve the performance of online attention-based encoder-decoder (AED) models.
no code implementations • 22 Dec 2019 • Chanwoo Kim, Mehul Kumar, Kwangyoun Kim, Dhananjaya Gowda
With the power function-based MUD, we apply a power-function based nonlinearity where power function coefficients are chosen to maximize the likelihood assuming that nonlinearity outputs follow the uniform distribution.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +1
no code implementations • 22 Dec 2019 • Chanwoo Kim, Sungsoo Kim, Kwangyoun Kim, Mehul Kumar, Jiyeon Kim, Kyungmin Lee, Changwoo Han, Abhinav Garg, Eunhyang Kim, Minkyoo Shin, Shatrughan Singh, Larry Heck, Dhananjaya Gowda
Our end-to-end speech recognition system built using this training infrastructure showed a 2. 44 % WER on test-clean of the LibriSpeech test set after applying shallow fusion with a Transformer language model (LM).