Title | ||
---|---|---|
Data augmentation using generative adversarial networks for robust speech recognition. |
Abstract | ||
---|---|---|
•This paper utilizes three different GANs for data augmentation to improve speech recognition under noise conditions.•The experiments show that out proposed data augmentation approaches can obtain the performance improvement under all noisy conditions, which have additive noise, channel distortion and reverberation.•With the proposed approach, we can use GAN to generate more training data under noisy conditions, which can be used in multi-condition training of acoustic modeling in robust speech recognition. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1016/j.specom.2019.08.006 | Speech Communication |
Keywords | Field | DocType |
Robust speech recognition,Generative adversarial networks,Conditional generative adversarial networks,Data augmentation,Very deep convolutional neural network | Reverberation,Pattern recognition,Computer science,Waveform,Communication channel,Speech recognition,Unsupervised learning,Artificial intelligence,Distortion,Test data generation,Performance improvement,Acoustic model | Journal |
Volume | ISSN | Citations |
114 | 0167-6393 | 2 |
PageRank | References | Authors |
0.35 | 0 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yanmin Qian | 1 | 295 | 44.44 |
Hu Hu | 2 | 2 | 0.69 |
Tian Tan | 3 | 99 | 6.27 |