iVector-based discriminative adaptation for automatic speech recognition

@article{Karafit2011iVectorbasedDA,
  title={iVector-based discriminative adaptation for automatic speech recognition},
  author={M. Karafi{\'a}t and L. Burget and P. Matejka and O. Glembek and J. Cernock{\'y}},
  journal={2011 IEEE Workshop on Automatic Speech Recognition & Understanding},
  year={2011},
  pages={152-157}
}
We presented a novel technique for discriminative feature-level adaptation of automatic speech recognition system. [...] Key Method To utilized iVectors for adaptation, Region Dependent Linear Transforms (RDLT) are discriminatively trained using MPE criterion on large amount of annotated data to extract the relevant information from iVectors and to compensate speech feature. The approach was tested on standard CTS data. We found it to be complementary to common adaptation techniques. On a well tuned RDLT system…Expand
58 Citations
I-vector dependent feature space transformations for adaptive speech recognition
  • 3
  • PDF
A comparative study of fMPE and RDLT approaches to LVCSR
  • 1
Transfer learning for automatic speech recognition systems
  • 2
Tied-state based discriminative training of context-expanded region-dependent feature transforms for LVCSR
  • 3
  • PDF
Lipreading with LipsID
Batch Normalization based Unsupervised Speaker Adaptation for Acoustic Models
  • Jiangyan Yi, J. Tao
  • Computer Science
  • 2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)
  • 2019
Adaptation of Deep Neural Networks
  • 6
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 20 REFERENCES
Recent progress on the discriminative region-dependent transform for speech feature extraction
  • 41
  • Highly Influential
  • PDF
A compact model for speaker-adaptive training
  • 582
  • PDF
Discriminative linear transforms for feature normalization and speaker adaptation in HMM estimation
  • 46
  • PDF
Minimum phoneme error based heteroscedastic linear discriminant analysis for speech recognition
  • B. Zhang, S. Matsoukas
  • Computer Science
  • Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005.
  • 2005
  • 22
  • Highly Influential
  • PDF
fMPE: discriminatively trained features for speech recognition
  • 320
  • PDF
Improvements in MLLR-Transform-based Speaker Recognition
  • 36
  • PDF
Discriminative linear transforms for speaker adaptation
  • 30
  • PDF
Discriminative Training for Large-Vocabulary Speech Recognition Using Minimum Classification Error
  • 610
  • PDF
Unsupervised discriminative adaptation using discriminative mapping transforms
  • K. Yu, M. Gales, P. Woodland
  • Computer Science
  • 2008 IEEE International Conference on Acoustics, Speech and Signal Processing
  • 2008
  • 13
  • Highly Influential
  • PDF
Discriminative speaker adaptation with conditional maximum likelihood linear regression
  • 73
  • PDF
...
1
2
...