Abstract | ||
---|---|---|
Recently, there is a research trend on ad-hoc microphone arrays. However, most research was conducted on simulated data. Although some datasets were collected with a small number of distributed devices, they were not synchronized which hinders the fundamental theoretical research on ad-hoc microphone arrays. To address this issue, this paper presents a synchronized speech corpus, named Libri-adhoc40, which collects the replayed Librispeech data from loudspeakers by ad-hoc microphone arrays of 40 strongly synchronized distributed nodes in a real office environment. Besides, to provide the evaluation target for speech frontend processing and other applications, we also recorded the replayed speech in an anechoic chamber. We trained several multi-device speech recognition systems on both the Libri-adhoc40 dataset and a simulated dataset. Experimental results demonstrate the validity of the proposed corpus which can be used as a benchmark to reflect the trend and difference of the models with different ad-hoc microphone arrays. The dataset is online available at https://github.com/ISmallFish/Libri-adhoc40. |
Year | Venue | DocType |
---|---|---|
2021 | 2021 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC) | Conference |
ISSN | Citations | PageRank |
2309-9402 | 0 | 0.34 |
References | Authors | |
0 | 11 |
Name | Order | Citations | PageRank |
---|---|---|---|
Shanzheng Guan | 1 | 0 | 0.34 |
Shupei Liu | 2 | 0 | 0.34 |
Junqi Chen | 3 | 0 | 0.68 |
Wenbo Zhu | 4 | 15 | 2.42 |
shengqiang li | 5 | 3 | 2.10 |
Xu Tan | 6 | 2 | 1.60 |
Ziye Yang | 7 | 0 | 0.68 |
Menglong Xu | 8 | 0 | 2.70 |
Yijiang Chen | 9 | 0 | 1.35 |
Jianyu Wang | 10 | 0 | 3.72 |
Xiao-Lei Zhang | 11 | 18 | 4.44 |