A time delay neural network architecture for efficient modeling of long temporal contexts

@inproceedings{Peddinti2015ATD,
  title={A time delay neural network architecture for efficient modeling of long temporal contexts},
  author={Vijayaditya Peddinti and Daniel Povey and Sanjeev Khudanpur},
  booktitle={INTERSPEECH},
  year={2015}
}
Recurrent neural network architectures have been shown to efficiently model long term temporal dependencies between acoustic events. However the training time of recurrent networks is higher than feedforward networks due to the sequential nature of the learning algorithm. In this paper we propose a time delay neural network architecture which models long term temporal dependencies with training times comparable to standard feed-forward DNNs. The network uses sub-sampling to reduce computation… CONTINUE READING
Highly Influential
This paper has highly influenced 28 other papers. REVIEW HIGHLY INFLUENTIAL CITATIONS
Highly Cited
This paper has 333 citations. REVIEW CITATIONS

Citations

Publications citing this paper.
Showing 1-10 of 232 extracted citations

Sequence Distillation for Purely Sequence Trained Acoustic Models

2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) • 2018
View 12 Excerpts
Highly Influenced

LIUM ASR systems for the 2016 Multi-Genre Broadcast Arabic challenge

2016 IEEE Spoken Language Technology Workshop (SLT) • 2016
View 10 Excerpts
Highly Influenced

Phonetic and Graphemic Systems for Multi-genre Broadcast Transcription

2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) • 2018
View 10 Excerpts
Highly Influenced

333 Citations

010020020152016201720182019
Citations per Year
Semantic Scholar estimates that this publication has 333 citations based on the available data.

See our FAQ for additional information.

References

Publications referenced by this paper.
Showing 1-10 of 32 references

Switchboard: telephone speech corpus for research and development

J. Godfrey, E. Holliman, J. McDaniel
Proceedings of the International Conference on Acoustics, Speech and Signal Processing (ICASSP), vol. 1, Mar 1992, pp. 517–520 vol.1. • 1992
View 5 Excerpts
Highly Influenced

Phoneme recognition using time-delay neural networks

IEEE Trans. Acoustics, Speech, and Signal Processing • 1989
View 4 Excerpts
Highly Influenced

An i-vector based time delay neural network architecture for far field recognition

V. Peddinti, G. Chen, D. Povey, S. Khudanpur
Proceedings of INTERSPEECH, 2015. [Online]. Available: http://www.danielpovey.com/files/ 2015 interspeech aspire.pdf • 2015
View 1 Excerpt

Adaptation of multilingual stacked bottle-neck neural network structure for new language

2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) • 2014
View 1 Excerpt

Deep Scattering Spectrum

IEEE Transactions on Signal Processing • 2014
View 1 Excerpt

Similar Papers

Loading similar papers…