Fetching the paper…

Reading the bibliography…

2012

Large Scale Language Modeling in Automatic Speech Recognition

Chelba, Ciprian, Bikel, Dan, Shugrina, Maria et al.

Original
PDF
Post on X

Understand

Large language models have been proven quite beneficial for a variety of automatic speech recognition tasks in Google.

We summarize results on Voice Search and a few YouTube speech transcription tasks to highlight the impact that one can expect from increasing both the amount of training data, and the size of the language model estimated from such data.
Depending on the task, availability and amount of training data used, language model size and amount of work and care put into integrating them in the lattice rescoring step we observe reductions in word error rate between 6% and 10% relative, for systems on a wide range of operating points between 17% and 52% word error rate.

Large Scale Language Modeling in Automatic Speech Recognition · Around

Built on

Information Extraction From Speech And Text
Frederick Jelinek, · 1997
Earlier work this paper cites.
“Large language models in machine translation,”
T. Brants, A. C. Popat, P. Xu, F. J. Och, and J. Dean, · 2007
Earlier work this paper cites.
“OpenFst: A general and efficient weighted finite-state transducer library,”
C. Allauzen, M. Riley, J. Schalkwyk, W. Skut, and M. Mohri, · 2007
Earlier work this paper cites.
“Boosted MMI for model and feature space discriminative training,”
D. Povey, D. Kanevsky, B. Kingsbury, B. Ramabhadran, G. Saon, and K. Visweswariah, · 2008
Earlier work this paper cites.

Similar

“Lattice-based minimum error rate training for statistical machine translation,”
W. Macherey, F. Och, I. Thayer, and J. Uszkoreit, · 2008
Cited alongside, same era.
“Back-off language model compression,”
B. Harb, C. Chelba, J. Dean, and S. Ghemawat, · 2009
Cited alongside, same era.
“A generalized composition algorithm for weighted finite-state transducers,”
C. Allauzen, J. Schalkwyk, and M. Riley, · 2009
Cited alongside, same era.
“An audio indexing system for election video material,”
C. Alberti, M. Bacchiani, A. Bezman, C. Chelba, A. Drofa, H. Liao, P. Moreno, T. Power, A. Sahuguet, M. Shugrina, and O. Siohan, · 2009
Cited alongside, same era.

Then

“Query language modeling for voice search,”
C. Chelba, J. Schalkwyk, T. Brants, V. Ha, B. Harb, W. Neveitt, C. Parada, and P. Xu, · 2010
Later among the works it cites.
“Application of pretrained deep neural networks to large vocabulary speech recognition,”
D. Jaitly, P. Nguyen, A. Senior, and V. Vanhoucke, · 2012
Closest in time.
“On-the-fly topic adaptation for youtube video transcription,”
K. Thadani, F. Biadsy, and D. Bikel, · 2012
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…