An investigation of augmenting speaker representations to improve speaker normalisation for DNN-based speech recognition
Authors:
- 1 April 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15206149,p. 4610-4613
Abstract
The conventional short-term interval features used by the Deep Neural Networks (DNNs) lack the ability to learn longer term information. This poses a challenge for training a speaker-independent (SI) DNN since the short-term features do not provide sufficient information for the DNN to estimate the real robust factors of speaker-level variations. The key to this problem is to obtain a sufficiently robust and informative speaker representation. This paper compares several speaker representations. Firstly, a DNN speaker classifier is used to extract the bottleneck features as the speaker representation, called the Bottleneck Speaker Vector (BSV). To further improve the robustness of this representation, a first-order Bottleneck Speaker Super Vector (BSSV) is also proposed, where the BSV is expanded into a super vector space by incorporating the phoneme posterior probabilities. Finally, a more fine-grain speaker representation based on the FMLLR-shifted features is examined. The experimental results on the WSJ0 and WSJ1 datasets show that the proposed speaker representations are useful in normalising the speaker effects for robust DNN-based automatic speech recognition. The best performance is achieved by augmenting both the BSSV and the FMLLR-shifted representations, yielding 10.0% - 15.3% relatively performance gains over the SI DNN baseline.Keywords
This publication has 9 references indexed in Scilit:
- Speaker adaptation of neural network acoustic models using i-vectors2013
- MLP-based factor analysis for tandem speech recognition2013
- Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research GroupsIEEE Signal Processing Magazine, 2012
- Feature engineering in Context-Dependent Deep Neural Networks for conversational speech transcription2011
- Speaker and Session Variability in GMM-Based Speaker VerificationIEEE Transactions on Audio, Speech, and Language Processing, 2007
- Modeling intra-speaker variability for speaker recognition2005
- Maximum likelihood linear transformations for HMM-based speech recognitionComputer Speech & Language, 1998
- Maximum likelihood linear regression for speaker adaptation of continuous density hidden Markov modelsComputer Speech & Language, 1995
- The design for the wall street journal-based CSR corpus1992