[1] [1] DDR3 4 2 CXP/Camera Link 7 626/840

[2] [2] FLASH 128 9 LVDS 2 016

[3] [3] PXIe SSD 2 970 6 TLK2711 1 200

[4] [4] DDR3 1 1 Camera Link 917

[5] [5] FLASH 16 1 TLK2711 -

[6] [6] FLASH 16 2 RS422/LVDS -

[7] [7] FLASH 128 24 LVDS 300

[8] [8] FLASH 128 24 LVDS 300

[9] [9] SATA SSD 8 192 8 ECL/LVDS 3 200/2 400

[10] [10] FLASH 1 30 LVDS 4 560

[11] [11] FLASH 0.5 17 LVDS - proposed NVMe SSD 8 192 8 QSFP 81 920

[23] [23] TAN Xu, ZHANG Xiaolei. Speech enhancement aided end-to-end multi-task learning for voice activity detection[EB/OL]. (2020-10-23). https://doi.org/10.48550/arXiv.2010.12484.

[24] [24] BARKER J,MARXER R,VINCENT E,et al. The third 'CHiME' speech separation and recognition challenge:dataset,task and baselines[C]// 2015 IEEE Workshop on Automatic Speech Recognition and Understanding(ASRU). Scottsdale,AZ,USA:IEEE, 2015:504-511.

[25] [25] VARGA A,STEENEKEN H J M. Assessment for automatic speech recognition:II. NOISEX-92:a database and an experiment to study the effect of additive noise on speech recognition systems[J]. Speech Communication, 1993,12(3):247-251.

[26] [26] PARK D S,CHAN W,ZHANG Y,et al. SpecAugment: a simple data augmentation method for automatic speech recognition[EB/ OL]. (2019-12-03). https://doi.org/10.48550/arXiv.1904.08779.