Efficient codebooks for fast and accurate low resource ASR systems

Authors:
Leila Zouari;Gérard Chollet
Affiliations:
GET - ENST/CNRS-LTCI, Département Traitement du Signal et des Images, 46 rue Barrault, 75634 Paris, France;GET - ENST/CNRS-LTCI, Département Traitement du Signal et des Images, 46 rue Barrault, 75634 Paris, France
Venue:
Speech Communication
Year:
2009

Citing 5
Cited 1

Improved modeling and efficiency for automatic transcription of Broadcast News

Speech Communication - Special issue on automatic transcription of broadcast news data
The bucket box intersection (BBI) algorithm for fast approximative evaluation of diagonal mixture Gaussians

ICASSP '96 Proceedings of the Acoustics, Speech, and Signal Processing, 1996. on Conference Proceedings., 1996 IEEE International Conference - Volume 02
Efficient speech recognition using subvector quantization and discrete-mixture HMMs

ICASSP '99 Proceedings of the Acoustics, Speech, and Signal Processing, 1999. on 1999 IEEE International Conference - Volume 02
Vector quantization for the efficient computation of continuous density likelihoods

ICASSP'93 Proceedings of the 1993 IEEE international conference on Acoustics, speech, and signal processing: speech processing - Volume II
A high-speed, low-resource ASR back-end based on custom arithmetic

IEEE Transactions on Audio, Speech, and Language Processing

Speech recognition on mobile devices

Mobile Multimedia Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Today, speech interfaces have become widely employed in mobile devices, thus recognition speed and resource consumption are becoming new metrics of Automatic Speech Recognition (ASR) performance. For ASR systems using continuous Hidden Markov Models (HMMs), the computation of the state likelihood is one of the most time consuming parts. In this paper, we propose novel multi-level Gaussian selection techniques to reduce the cost of state likelihood computation. These methods are based on original and efficient codebooks. The proposed algorithms are evaluated within the framework of a large vocabulary continuous speech recognition task.