Abstract
The Single Multimodal Android Service for HCI (SMASH) framework implements an automated language data generation algorithm to support high-accuracy, efficient, always-listening voice command recognition using the Carnegie Mellon University (CMU) PocketSphinx n-gram speech recognizer. SMASH injects additional language data into the language model generation process to augment the orthographies extracted from the input voice command grammar. This additional data allows for a larger variety of potential outcomes, and greater phonetic distance between outcomes, within the generated language model, resulting in more consistent probability scores for in-grammar utterances, and fewer false positives from out-of-grammar (OOG) utterances.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Intent | Android Developers. https://developer.android.com/reference/android/content/intent.html
JSpeech Grammar Format. https://www.w3.org/TR/jsgf/
CMU Sphinx. http://cmusphinx.sourceforge.net
Services | Android Developers. https://developer.android.com/guide/components/services.html
JSGFGrammar (Sphinx-4). http://cmusphinx.sourceforge.net/doc/sphinx4/edu/cmu/sphinx/jsgf/JSGFGrammar.html
Bellegarda, J.R.: An overview of statistical language model adaptation. In: ITRW on Adaptation Methods for Speech Recognition. Sophia Antipolis, France (2001)
Random Word Generator. https://randomwordgenerator.com
The ARPA-MIT LM format. http://www1.icsi.berkeley.edu/Speech/docs/HTKBook3.2/node213_mn.html
The CMU Pronouncing Dictionary. http://www.speech.cs.cmu.edu/cgi-bin/cmudict
Clarkson, P.: The CMU-Cambridge Statistical Language Modeling Toolkit v2. http://www.speech.cs.cmu.edu/SLM/toolkit_documentation.html
FreeTTS Programmer’s Guide. http://freetts.sourceforge.net/docs/ProgrammerGuide.html
Black, A.W., Lenzo, K., Pagel, V.: Issues in building general letter-to-sound rules. In: Proceedings of ECSA Workshop on Speech Synthesis, pp. 77–80, Australia (1998)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG (outside the USA)
About this paper
Cite this paper
Yaeger, D., Bubeck, C. (2018). Efficient Language Model Generation Algorithm for Mobile Voice Commands. In: Ahram, T., Karwowski, W. (eds) Advances in Human Factors, Software, and Systems Engineering. AHFE 2017. Advances in Intelligent Systems and Computing, vol 598. Springer, Cham. https://doi.org/10.1007/978-3-319-60011-6_11
Download citation
DOI: https://doi.org/10.1007/978-3-319-60011-6_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-60010-9
Online ISBN: 978-3-319-60011-6
eBook Packages: EngineeringEngineering (R0)