Speech recognition modeling advances for mobile voice search

Enrico Bocchieri, Diamantino Caseiro, Dimitrios Dimitriadis

Abstract

This paper reports on the development and advances in automatic speech recognition for the AT&T Speak4it <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> voice-search application. With Speak4it as real-life example, we show the effectiveness of acoustic model (AM) and language model (LM) estimation (adaptation and training) on relatively small amounts of application field-data. We then introduce algorithmic improvements concerning the use of sentence length in LM, of non-contextual features in AM decision-trees, and of the Teager energy in the acoustic front-end. The combination of these algorithms, integrated into the AT&T Watson recognizer, yields substantial accuracy improvements. LM and AM estimation on field-data samples increases the word accuracy from 66.4% to 77.1%, a relative word error reduction of 32%. The algorithmic improvements increase the accuracy to 79.7%, an additional 11.3% relative error reduction.
BibTeX
@inproceedings{bocchieri2011speech,
  author = {Enrico Bocchieri and Diamantino Caseiro and Dimitrios Dimitriadis},
  title = {{Speech recognition modeling advances for mobile voice search}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2011},
  doi = {10.1109/icassp.2011.5947451},
}

Loading related content...

Figure