Abstract | ||
---|---|---|
Recently, speaker adaptation methods in deep neural networks (DNNs) have been widely studied for automatic speech recognition. However, almost all adaptation methods for DNNs have to consider various heuristic conditions such as mini-batch sizes, learning rate scheduling, stopping criteria, and initialization conditions because of the inherent property of a stochastic gradient descent (SGD)-based training process. Unfortunately, those heuristic conditions are hard to be properly tuned. To alleviate those difficulties, in this paper, we propose a least squares regression -based speaker adaptation method in a DNN framework utilizing posterior mean of each class. Also, we show how the proposed method can provide a unique solution which is quite easy and fast to calculate without SGD. The proposed method was evaluated in the TED-LIUM corpus. Experimental results showed that the proposed method achieved up to a 4.6% relative improvement against a speaker independent DNN. In addition, we report further performance improvement of the proposed method with speaker-adapted features. |
Year | DOI | Venue |
---|---|---|
2017 | 10.21437/Interspeech.2017-783 | 18TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2017), VOLS 1-6: SITUATED INTERACTION |
Keywords | Field | DocType |
deep neural network, speaker adaptation, class-dependent posterior mean, deep least squares regression | Least squares,Pattern recognition,Computer science,Speech recognition,Artificial intelligence,Speaker adaptation | Conference |
ISSN | Citations | PageRank |
2308-457X | 0 | 0.34 |
References | Authors | |
12 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Younggwan Kim | 1 | 17 | 6.11 |
Hyungjun Lim | 2 | 31 | 7.66 |
Jahyun Goo | 3 | 0 | 1.35 |
Hoirin Kim | 4 | 0 | 1.01 |