English

The Microsoft 2016 Conversational Speech Recognition System

Computation and Language 2022-02-24 v2 Audio and Speech Processing

Abstract

We describe Microsoft's conversational speech recognition system, in which we combine recent developments in neural-network-based acoustic and language modeling to advance the state of the art on the Switchboard recognition task. Inspired by machine learning ensemble techniques, the system uses a range of convolutional and recurrent neural networks. I-vector modeling and lattice-free MMI training provide significant gains for all acoustic model architectures. Language model rescoring with multiple forward and backward running RNNLMs, and word posterior-based system combination provide a 20% boost. The best single system uses a ResNet architecture acoustic model with RNNLM rescoring, and achieves a word error rate of 6.9% on the NIST 2000 Switchboard task. The combined system has an error rate of 6.2%, representing an improvement over previously reported results on this benchmark task.

Keywords

Cite

@article{arxiv.1609.03528,
  title  = {The Microsoft 2016 Conversational Speech Recognition System},
  author = {W. Xiong and J. Droppo and X. Huang and F. Seide and M. Seltzer and A. Stolcke and D. Yu and G. Zweig},
  journal= {arXiv preprint arXiv:1609.03528},
  year   = {2022}
}
R2 v1 2026-06-22T15:47:29.895Z