2017 Hands-free Speech Communications and Microphone Arrays (HSCMA)最新文献_第2页

A unified deep modeling approach to simultaneous speech dereverberation and recognition for the reverb challenge 针对混响挑战的语音同步去混响和识别的统一深度建模方法

2017 Hands-free Speech Communications and Microphone Arrays (HSCMA) Pub Date : 2017-03-01 DOI: 10.1109/HSCMA.2017.7895557

Bo Wu, Kehuang Li, Zhen Huang, S. Siniscalchi, Minglei Yang, Chin-Hui Lee

{"title":"A unified deep modeling approach to simultaneous speech dereverberation and recognition for the reverb challenge","authors":"Bo Wu, Kehuang Li, Zhen Huang, S. Siniscalchi, Minglei Yang, Chin-Hui Lee","doi":"10.1109/HSCMA.2017.7895557","DOIUrl":"https://doi.org/10.1109/HSCMA.2017.7895557","url":null,"abstract":"We propose a unified deep neural network (DNN) approach to achieve both high-quality enhanced speech and high-accuracy automatic speech recognition (ASR) simultaneously on the recent REverberant Voice Enhancement and Recognition Benchmark (RE-VERB) Challenge. These two goals are accomplished by two proposed techniques, namely DNN-based regression to enhance reverberant and noisy speech, followed by DNN-based multi-condition training that takes clean-condition, multi-condition and enhanced speech all into consideration. We first report on superior objective measures in enhanced speech to those listed in the 2014 REVERB Challenge Workshop. We then show that in clean-condition training, we obtain the best word error rate (WER) of 13.28% on the 1-channel REVERB simulated evaluation data with the proposed DNN-based pre-processing scheme. Similarly we attain a competitive single-system WER of 8.75% with the proposed multi-condition training strategy and the same less-discriminative log power spectrum features used in the enhancement stage. Finally by leveraging upon joint training with more discriminative ASR features and improved neural network based language models a state-of-the-art WER of 4.46% is attained with a single ASR system, and single-channel information. Another state-of-the-art WER of 4.10% is achieved through system combination.","PeriodicalId":337673,"journal":{"name":"2017 Hands-free Speech Communications and Microphone Arrays (HSCMA)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115149066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 6

Deep bi-directional long short-term memory based speech enhancement for wind noise reduction 基于深度双向长短期记忆的语音增强降噪研究

2017 Hands-free Speech Communications and Microphone Arrays (HSCMA) Pub Date : 2017-03-01 DOI: 10.1109/HSCMA.2017.7895558

Jinkyu Lee, Keulbit Kim, T. Shabestary, Hong-Goo Kang

引用次数: 16

Multigrid reconstruction of sound fields using moving microphones 移动传声器声场的多网格重建

2017 Hands-free Speech Communications and Microphone Arrays (HSCMA) Pub Date : 2017-03-01 DOI: 10.1109/HSCMA.2017.7895588

Fabrice Katzberg, Radoslaw Mazur, M. Maass, P. Koch, A. Mertins

引用次数: 3

A unified framework for multiple arrays on a robot and application to sound localization 机器人上多个阵列的统一框架及声音定位应用

2017 Hands-free Speech Communications and Microphone Arrays (HSCMA) Pub Date : 2017-03-01 DOI: 10.1109/HSCMA.2017.7895563

Lior Madmoni, Hendrik Barfuss, B. Rafaely, Walter Kellermann

引用次数: 0

A study on search grid points for data-driven 3-D beamsteering 数据驱动的三维波束导向搜索网格点研究

2017 Hands-free Speech Communications and Microphone Arrays (HSCMA) Pub Date : 2017-03-01 DOI: 10.1109/HSCMA.2017.7895551

JeeSok Lee, Soo-Whan Chung, Min-Seok Choi, Hong-Goo Kang

引用次数: 0

Sound field separation in a mixed acoustic environment using a sparse array of higher order spherical microphones 高阶球形传声器稀疏阵列在混合声环境中的声场分离

2017 Hands-free Speech Communications and Microphone Arrays (HSCMA) Pub Date : 2017-03-01 DOI: 10.1109/HSCMA.2017.7895580

Abdullah Fahim, P. Samarasinghe, T. Abhayapala

引用次数: 15

Speaker tracking in reverberant environments using multiple directions of arrival 混响环境中使用多个到达方向的说话人跟踪

2017 Hands-free Speech Communications and Microphone Arrays (HSCMA) Pub Date : 2017-03-01 DOI: 10.1109/HSCMA.2017.7895568

C. Evers, B. Rafaely, P. Naylor

{"title":"Speaker tracking in reverberant environments using multiple directions of arrival","authors":"C. Evers, B. Rafaely, P. Naylor","doi":"10.1109/HSCMA.2017.7895568","DOIUrl":"https://doi.org/10.1109/HSCMA.2017.7895568","url":null,"abstract":"Accurate estimation of the Direction of Arrival (DOA) of a sound source is an important prerequisite for a wide range of acoustic signal processing applications. However, in enclosed environments, early reflections and late reverberation often lead to localization errors. Recent work demonstrated that improved robustness against reverberation can be achieved by clustering only the DOAs from direct-path bins in the short-term Fourier transform of a speech signal of several seconds duration from a static talker. Nevertheless, for moving talkers, short blocks of at most several hundred milliseconds are required to capture the spatio-temporal variation of the source direction. Processing of short blocks of data in reverberant environment can lead to clusters whose centroids correspond to spurious DOAs away from the source direction. We therefore propose in this paper a novel multi-detection source tracking approach that estimates the smoothed trajectory of the source DOAs. Results for realistic room simulations validate the proposed approach and demonstrate significant improvements in estimation accuracy compared to single-detection tracking.","PeriodicalId":337673,"journal":{"name":"2017 Hands-free Speech Communications and Microphone Arrays (HSCMA)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125487256","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3

An extended experimental investigation of DNN uncertainty propagation for noise robust ASR 噪声鲁棒ASR中深度神经网络不确定性传播的扩展实验研究

2017 Hands-free Speech Communications and Microphone Arrays (HSCMA) Pub Date : 2017-03-01 DOI: 10.1109/HSCMA.2017.7895555

K. Nathwani, J. A. Morales-Cordovilla, S. Sivasankaran, I. Illina, E. Vincent

{"title":"An extended experimental investigation of DNN uncertainty propagation for noise robust ASR","authors":"K. Nathwani, J. A. Morales-Cordovilla, S. Sivasankaran, I. Illina, E. Vincent","doi":"10.1109/HSCMA.2017.7895555","DOIUrl":"https://doi.org/10.1109/HSCMA.2017.7895555","url":null,"abstract":"Automatic speech recognition (ASR) in noisy environments remains a challenging goal. Recently, the idea of estimating the uncertainty about the features obtained after speech enhancement and propagating it to dynamically adapt deep neural network (DNN) based acoustic models has raised some interest. However, the results in the literature were reported on simulated noisy datasets for a limited variety of uncertainty estimators. We found that they vary significantly in different conditions. Hence, the main contribution of this work is to assess DNN uncertainty decoding performance for different data conditions and different uncertainty estimation/propagation techniques. In addition, we propose a neural network based uncertainty estimator and compare it with other uncertainty estimators. We report detailed ASR results on the CHiME-2 and CHiME-3 datasets. We find that, on average, uncertainty propagation provides similar relative improvement on real and simulated data and that the proposed uncertainty estimator performs significantly better than the one in [1]. We also find that the improvement is consistent, but it depends on the signal-to-noise ratio (SNR) and the noise environment.","PeriodicalId":337673,"journal":{"name":"2017 Hands-free Speech Communications and Microphone Arrays (HSCMA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115844520","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 10

Towards real-time source counting by estimation of coherent-to-diffuse ratios from ad-hoc microphone array recordings 对实时源计数的估计相干漫射比从特设麦克风阵列记录

2017 Hands-free Speech Communications and Microphone Arrays (HSCMA) Pub Date : 2017-03-01 DOI: 10.1109/HSCMA.2017.7895582

Shahab Pasha, Jacob Donley, C. Ritz, Y. Zou

引用次数: 7

Online meeting recognition in noisy environments with time-frequency mask based MVDR beamforming 基于时频掩模的MVDR波束形成在噪声环境下的在线会议识别

2017 Hands-free Speech Communications and Microphone Arrays (HSCMA) Pub Date : 2017-03-01 DOI: 10.1109/HSCMA.2017.7895553

S. Araki, N. Ito, Marc Delcroix, A. Ogawa, K. Kinoshita, T. Higuchi, Takuya Yoshioka, Dung T. Tran, Shigeki Karita, T. Nakatani

{"title":"Online meeting recognition in noisy environments with time-frequency mask based MVDR beamforming","authors":"S. Araki, N. Ito, Marc Delcroix, A. Ogawa, K. Kinoshita, T. Higuchi, Takuya Yoshioka, Dung T. Tran, Shigeki Karita, T. Nakatani","doi":"10.1109/HSCMA.2017.7895553","DOIUrl":"https://doi.org/10.1109/HSCMA.2017.7895553","url":null,"abstract":"This paper addresses our new online meeting recognition prototype, which works even in noisy environments. For speech enhancement, we employ a mask-based minimum variance distortionless response (MVDR) beamformer, which has recently shown to be a successful front-end for a state-of-the-art deep neural network (DNN)-based automatic speech recognition (ASR) system. To ensure more accurate and computationally efficient mask estimation for estimating steering vectors of a MVDR beamformer, which is challenging especially in an adverse environment, this work employs a probabilistic spatial dictionary. The dictionary consists of a pre-trained probability distribution of source location features for each of speaker location candidates. This dictionary significantly simplifies the problem to be solved, and therefore, realizes an accurate and efficient mask estimation. Our new prototype worked with stability with a latency of 3–5 seconds in a real exhibition situation. The quantitative evaluation results will be reported.","PeriodicalId":337673,"journal":{"name":"2017 Hands-free Speech Communications and Microphone Arrays (HSCMA)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123104492","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 15