Search Results - "Chongjia Ni"

Preventing Early Endpointing for Online Automatic Speech Recognition by Zhao, Yingzhu, Ni, Chongjia, Leung, Cheung-Chi, Joty, Shafiq, Chng, Eng Siong, Ma, Bin

“…With the recent development of end-to-end models in speech recognition, there have been more interests in adapting these models for online speech recognition…”
Get full text

Modification on LSA speech enhancement for speech recognition by Chang Huai You, Bin Ma, Chongjia Ni

“…Speech recognition performance deteriorates in face of unknown noise. Speech enhancement offers a solution by reducing the noise in speech at runtime. However,…”
Get full text

Efficient methods to train multilingual bottleneck feature extractors for low resource keyword search by Chongjia Ni, Cheung-Chi Leung, Lei Wang, Chen, Nancy F., Bin Ma

“…Training a bottleneck feature (BNF) extractor with multilingual data has been common in low resource keyword search. In a low resource application, the amount…”
Get full text

Unsupervised data selection and word-morph mixed language model for tamil low-resource keyword search by Chongjia Ni, Cheung-Chi Leung, Lei Wang, Chen, Nancy F., Bin Ma

“…This paper considers an unsupervised data selection problem for the training data of an acoustic model and the vocabulary coverage of a keyword search system…”
Get full text

From English pitch accent detection to Mandarin stress detection, where is the difference? by Ni, Chongjia, Liu, Wenju, Xu, Bo

“…► The classifier combination method, which is the combination of boosting classification and regression tree and conditional random fields, is employed to…”
Get full text

A keyword-aware grammar framework for LVCSR-based spoken keyword search by I-Fan Chen, Chongjia Ni, Boon Pang Lim, Chen, Nancy F., Chin-Hui Lee

“…In this paper, we proposed a method to realize the recently developed keyword-aware grammar for LVCSR-based keyword search using weight finite-state automata…”
Get full text

Cross-lingual deep neural network based submodular unbiased data selection for low-resource keyword search by Chongjia Ni, Cheung-Chi Leung, Lei Wang, Haibo Liu, Feng Rao, Li Lu, Chen, Nancy F., Bin Ma, Haizhou Li

“…In this paper, we propose a cross-lingual deep neural network (DNN) based submodular unbiased data selection approach for low-resource keyword search (KWS). A…”
Get full text

Submodular data selection with acoustic and phonetic features for automatic speech recognition by Chongjia Ni, Lei Wang, Haibo Liu, Cheung-Chi Leung, Li Lu, Bin Ma

“…In this paper, we propose to use acoustic feature based submodular function optimization to select a subset of untranscribed data for manual transcription, and…”
Get full text

Long short-term memory recurrent neural network based segment features for music genre classification by Jia Dai, Shan Liang, Wei Xue, Chongjia Ni, Wenju Liu

“…In the conventional frame feature based music genre classification methods, the audio data is represented by independent frames and the sequential nature of…”
Get full text

Investigate automatic speech recognition and keyword search for very low-resource language by Chongjia Ni, Bin Ma

“…In this paper, pronunciation lexicon, multi-lingual bottleneck features, semi-supervised learning, and data selection are investigated to help to improve the…”
Get full text

Investigation of using different Chinese word segmentation standards and algorithms for automatic speech recognition by Chongjia Ni, Cheung-Chi Leung

“…Chinese word segmentation (CWS) is a necessary step in Mandarin Chinese automatic speech recognition (ASR), and it has an impact on the results of ASR…”
Get full text

A novel codebook representation method and encoding strategy for bag-of-words based acoustic event classification by Jia Dai, Chongjia Ni, Wei Xue, Wenju Liu

“…The bag-of-words (BoW) model has been widely used for acoustic event classification (AEC). The performance of the BoW based AEC model is much influenced by…”
Get full text

A novel keyword+LVCSR-filler based grammar network representation for spoken keyword search by I-Fan Chen, Chongjia Ni, Boon Pang Lim, Chen, Nancy F., Chin-Hui Lee

“…A novel spoken keyword search grammar representation framework is proposed to combine the advantages of conventional keyword-filler based keyword search (KWS)…”
Get full text

Multiple time-span feature fusion for deep neural network modeling by Chongjia Ni, Chen, Nancy F., Bin Ma

“…In this paper, we exploit long term information from multiple time-spans for automatic speech recognition. The multiple time-span information is encoded into…”
Get full text

De'hubert: Disentangling Noise in a Self-Supervised Model for Robust Speech Recognition by Ng, Dianwen, Zhang, Ruixi, Yip, Jia Qi, Yang, Zhao, Ni, Jinjie, Zhang, Chong, Ma, Yukun, Ni, Chongjia, Chng, Eng Siong, Ma, Bin

“…Existing self-supervised pre-trained speech models have offered an effective way to leverage massive unannotated corpora to build good automatic speech…”
Get full text

Contrastive Speech Mixup for Low-Resource Keyword Spotting by Ng, Dianwen, Zhang, Ruixi, Yip, Jia Qi, Zhang, Chong, Ma, Yukun, Nguyen, Trung Hieu, Ni, Chongjia, Chng, Eng Siong, Ma, Bin

“…Most of the existing neural-based models for keyword spotting (KWS) in smart devices require thousands of training samples to learn a decent audio…”
Get full text

SPGM: Prioritizing Local Features for Enhanced Speech Separation Performance by Yip, Jia Qi, Zhao, Shengkui, Ma, Yukun, Ni, Chongjia, Zhang, Chong, Wang, Hao, Nguyen, Trung Hieu, Zhou, Kun, Ng, Dianwen, Chng, Eng Siong, Ma, Bin

“…Dual-path is a popular architecture for speech separation models (e.g. Sepformer) which splits long sequences into overlapping chunks for its intra- and…”
Get full text

Are Soft Prompts Good Zero-Shot Learners for Speech Recognition? by Ng, Dianwen, Zhang, Chong, Zhang, Ruixi, Ma, Yukun, Ritter-Gutierrez, Fabian, Nguyen, Trung Hieu, Ni, Chongjia, Zhao, Shengkui, Chng, Eng Siong, Ma, Bin

“…Large self-supervised pre-trained speech models require computationally expensive fine-tuning for downstream tasks. Soft prompt tuning offers a simple…”
Get full text

MossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech Separation by Zhao, Shengkui, Ma, Yukun, Ni, Chongjia, Zhang, Chong, Wang, Hao, Nguyen, Trung Hieu, Zhou, Kun, Yip, Jia Qi, Ng, Dianwen, Ma, Bin

“…Our previously proposed MossFormer has achieved promising performance in monaural speech separation. However, it predominantly adopts a self-attention-based…”
Get full text

Independent Language Modeling Architecture for End-To-End ASR by Pham, Van Tung, Xu, Haihua, Khassanov, Yerbolat, Zeng, Zhiping, Chng, Eng Siong, Ni, Chongjia, Ma, Bin, Li, Haizhou

“…The attention-based end-to-end (E2E) automatic speech recognition (ASR) architecture allows for joint optimization of acoustic and language models within a…”
Get full text