A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2023; you can also visit the original URL.
The file type is application/pdf.
Filters
SA-Paraformer: Non-autoregressive End-to-End Speaker-Attributed ASR
[article]
2023
arXiv
pre-print
Preserved Fulltext
Joint modeling of multi-speaker ASR and speaker diarization has recently shown promising results in speaker-attributed automatic speech recognition (SA-ASR).Although being able to obtain state-of-the-art (SOTA) performance, most of the studies are based on an autoregressive (AR) decoder which generates tokens one-by-one and results in a large real-time factor (RTF). To speed up inference, we introduce a recently proposed non-autoregressive model Paraformer as an acoustic model in the SA-ASR
arXiv:2310.04863v1
fatcat:22evywa5pzc5fawvcla7paaqay
more »
... l.Paraformer uses a single-step decoder to enable parallel generation, obtaining comparable performance to the SOTA AR transformer models. Besides, we propose a speaker-filling strategy to reduce speaker identification errors and adopt an inter-CTC strategy to enhance the encoder's ability in acoustic modeling. Experiments on the AliMeeting corpus show that our model outperforms the cascaded SA-ASR model by a 6.1% relative speaker-dependent character error rate (SD-CER) reduction on the test set. Moreover, our model achieves a comparable SD-CER of 34.8% with only 1/10 RTF compared with the SOTA joint AR SA-ASR model.
BA-SOT: Boundary-Aware Serialized Output Training for Multi-Talker ASR
[article]
2023
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2023; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2023 pre-print arXiv:2305.13716v1 fatcat:afav5tj5nfhfnazffo7wjiv5oyOther Versions
The recently proposed serialized output training (SOT) simplifies multi-talker automatic speech recognition (ASR) by generating speaker transcriptions separated by a special token. However, frequent speaker changes can make speaker change prediction difficult. To address this, we propose boundary-aware serialized output training (BA-SOT), which explicitly incorporates boundary knowledge into the decoder via a speaker change detection task and boundary constraint loss. We also introduce a
arXiv:2305.13716v3
fatcat:vrlwsrsgjjfotdkib5l7c4hgrq
more »
... ge connectionist temporal classification (CTC) strategy that incorporates token-level SOT CTC to restore temporal context information. Besides typical character error rate (CER), we introduce utterance-dependent character error rate (UD-CER) to further measure the precision of speaker change prediction. Compared to original SOT, BA-SOT reduces CER/UD-CER by 5.1%/14.0%, and leveraging a pre-trained ASR model for BA-SOT model initialization further reduces CER/UD-CER by 8.4%/19.9%.
Cross-linking of poly(dimethylaminoethyl methacrylate) by phytic acid: pH-responsive adsorbent for high-efficiency removal of cationic and anionic dyes
2020
RSC Advances
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
The adsorbent PADG based on phytic acid and DMAEMA was synthesized and tested, which is pH-sensitive and shows high adsorption capacities for anionic and cationic dyes.
doi:10.1039/c9ra09391e
pmid:35495251
pmcid:PMC9049133
fatcat:gcdszuqqavbhvce2jvihgh4mmu
ElectronMomentumSpectroscopy for Saturated Alkanes CnH2n+2(n=4-6)
2010
Wuli huaxue xuebao
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Acknowledgments： Oneoftheauthors,YANGZe鄄Jin(ZY),thanks SwinburneUniversityofTechnology(SUT,Australia)forhospitality. ZYcompleteddoctoralthesisresearchatSUTsupervisedbyProfessor WANGFeng. ...
No.9
Isomerindependenceoftherelativeintensityof theinnermostvalenceorbitals YANGZe鄄Jin etal.： ElectronMomentumSpectroscopyforSaturatedAlkanesC n H 2 n +2 (n=4-6) tane跃iso-butane, n-pentane跃iso-pentane ...
Isomerdependenceoftherelativeintensityof valenceorbitalsofalkane Selectedelectronorbitalmomentumdistributionsfor n鄄butane and iso 鄄 butaneareshowninFig.3tounderstandthecarbon chainbranchinginbutane.Theselectedrepresentativeorbitals No.9 YANGZe ...
doi:10.3866/pku.whxb20100924
fatcat:j5ca4tklkvb3jk3bibaqzemd5a
BA-SOT: Boundary-Aware Serialized Output Training for Multi-Talker ASR
2023
INTERSPEECH 2023
unpublished
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2023; you can also visit the original URL.
The file type is application/pdf.
The recently proposed serialized output training (SOT) simplifies multi-talker automatic speech recognition (ASR) by generating speaker transcriptions separated by a special token. However, frequent speaker changes can make speaker change prediction difficult. To address this, we propose boundaryaware serialized output training (BA-SOT), which explicitly incorporates boundary knowledge into the decoder via a speaker change detection task and boundary constraint loss. We also introduce a
doi:10.21437/interspeech.2023-1521
fatcat:os4qk5neavf5zi2fbyfsx4evcq
more »
... e connectionist temporal classification (CTC) strategy that incorporates token-level SOT CTC to restore temporal context information. Besides typical character error rate (CER), we introduce utterance-dependent character error rate (UD-CER) to further measure the precision of speaker change prediction. Compared to original SOT, BA-SOT reduces CER/UD-CER by 5.1%/14.0%, and leveraging a pre-trained ASR model for BA-SOT model initialization further reduces CER/UD-CER by 8.4%/19.9%.
The NPU System for DASR Task of CHiME-7 Challenge
2023
7th International Workshop on Speech Processing in Everyday Environments (CHiME 2023)
unpublished
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2023; you can also visit the original URL.
The file type is application/pdf.
This study describes the NPU system for the Distant Automatic Speech Recognition (DASR) task of the CHiME-7 Challenge. Specifically, two attention-based channel selection modules are introduced to automatically select the most advantageous channel subset from multiple signal channels. Furthermore, we incorporate additional spatial features during the cross-channel attention, which guides the model to capture the desired signals while suppressing the interference sources. It is noteworthy that
doi:10.21437/chime.2023-12
fatcat:kcsurz6bmnhnbkr3yi6yezcrui
more »
... ese enhancements solely pertain to the ASR model, with no modifications made to the speaker diarization (SD). Our approach achieves a Macro diarization attributed word error rate (DA-WER) of 22.28% on CHiME-7 dev sets with oracle diarization and 41.04% on CHiME-7 dev sets with baseline SD results.
Crustal structure beneath the Qilian Orogen Zone from multiscale seismic tomography
2019
Earth and Planetary Physics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
The Alxa, Ordos, and Yangze show very high-velocity anomalies. ...
Guo B et al.: Crustal structure of Qilian Orogen ...
doi:10.26464/epp2019025
fatcat:pxfwckys3nevlksxb2dzsdmg74
Chromosome variation in the genus Pinellia (Araceae) in China and Japan
2005
Botanical journal of the Linnean Society
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Based on chromosome studies of 11 populations of P. ternata , together with 12 populations reported in previous studies, the lower reaches of the Yangze River are identified as its centre of origin. ...
Guo & Zhuang (1988)
26
13
?
Li (1995)
26
13
China, Hubei (Cult.) ...
Guo & X. L. Liu has been combined with P. ternata (Thunb.) Breit. (Yi, 2002) , it now includes only seven perennial herbaceous species. ...
doi:10.1111/j.1095-8339.2005.00381.x
fatcat:ghc6cf5qjzezpgr3ajzoy5quzy
Clinical analysis of bevacizumab targeting therapy in treating early colorectal carcinoma after operation
2017
Oncology Letters
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
micropump in ivgtt for 46 h; one course was for two weeks and at least three courses were needed; 5 mg/kg bevacizumab was administered in the observation group (bevacizumab, Avastin ® , 100 mg/4 ml; Yangze ...
doi:10.3892/ol.2017.6087
pmid:28599469
pmcid:PMC5452938
fatcat:quekutxiibeyfpttvbnlvy3h3u
Genetic diversity among red swamp crayfish (Procambarus clarkii) populations in the middle and lower reaches of the Yangtze River based on AFLP markers
2013
Genetics and Molecular Research
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
We conclude that there is high genetic differentiation among crayfish in the middle and lower reaches of the Yangze River. ...
At present, although it is widely distributed in the middle and lower reaches of the Yangze River basin, little is known about its population genetics and geographic distribution in China. ...
For example, the PY a population exhibits a large seasonal water level fluctuation during the rainy season from April/ May to October (Guo et al., 2005) . ...
doi:10.4238/2013.march.13.8
pmid:23546963
fatcat:eddfhn6ljjdrnk7w43esbgk7ba
Table of Contents
2021
2021 4th International Conference on Advanced Electronic Materials, Computers and Software Engineering (AEMCSE)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
, China) , and Mingshi Liu (Beijing Institute of Technology, China) Traction Network Resonance Suppression Strategy Based on Auxiliary Filter Winding of Traction Transformer 249 Haiteng Wang (CRRC Yangze ...
Gao
(Beijing Institute of Technology, China), and Jingliang Lv (Tsinghua
University, China)
A New Algorithm for Robot Path Planning Based on Automatic Shunting of Ants and Particle Swarms 380 Guo ...
doi:10.1109/aemcse51986.2021.00004
fatcat:hibtkylopfe5vasdfwsnbb5guy
Determinations and Distributions of Acid Volatile Sulfide (AVS) in the Sediments of Poyang Lake, China
2017
DEStech Transactions on Environment Energy and Earth Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Experimental Section
Site Description Poyang Lake, which is the largest freshwater lake in China is located in the middle and lower reaches of Yangze River. ...
doi:10.12783/dteees/edep2017/15540
fatcat:soo3vwvpsreofgicdbeajyc5te
Impact of diurnal variability and meteorological factors on the PM 2.5 - AOD relationship: Implications for PM 2.5 remote sensing
2017
Environmental Pollution
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
., 2014a; Guo et al., 2016a) . ...
Besides, morning PM 2.5 peak dominates the Yangze River Delta region (YRD in Fig. 1b ), with amplitude lying between those of NCP and PRD. ...
doi:10.1016/j.envpol.2016.11.043
pmid:27889085
fatcat:q5eim4qrk5grrfds2v4kuhqgvu
Identification of the source of A (H10N8) virus causing human infection
2015
Infection, Genetics and Evolution
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
The recent emergence of H7N9 low pathogenic avian influenza viruses in Yangze Delta has six genes derived from H9N2 viruses, and mixed infections with H7N9 and H9N2 were very common (Gao et al., 2013; ...
The first H9N2 low pathogenic avian influenza virus was initially isolated from domestic poultry in 1994 (Guo et al., 2003) , and has since been found to be endemic in domestic poultry in China (Li et ...
doi:10.1016/j.meegid.2014.12.026
pmid:25550151
pmcid:PMC4838479
fatcat:p7e7x66gqvbcxgygk22tateqgy
Spatial-temporal distribution and impact factors of irrigation water use efficiency in the grain production of China
2018
International Journal of Agricultural and Biological Engineering
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
HH PAMs are distributed in north of the Yangze River and the plain of the middle and lower reaches of the Yellow River. ...
marginal water productivity for 31 PAMs in 1998, 2005 and 2010
Figure 4 4 Distribution of the marginal water productivity and irrigation proportion of arable land (IP) in China
Biographies: Xiangping Guo ...
doi:10.25165/j.ijabe.20181105.3588
fatcat:63cfm5wzfradlk7thwou6bhuim
« Previous
Showing results 1 — 15 out of 62 results