Improved noise-robustness in distributed speech recognition via perceptually-weighted vector quantisation of filterbank energies

No Thumbnail Available
File version
Author(s)
So, S
Paliwal, KK
Griffith University Author(s)
Primary Supervisor
Other Supervisors
Editor(s)

Luís Oliveira

Date
2005
Size
File type(s)
Location

Lisbon, Portugal

License
Abstract

In this paper, we examine a coding scheme for quantising feature vectors in a distributed speech recognition environment that is more robust to noise. It consists of a vector quantiser that operates on the logarithmic filterbank energies (LFBEs). Through the use of a perceptually-weighted Euclidean distance measure, which emphasises the LFBEs that represent the spectral peaks, the vector quantiser codebook provides /emph{a priori} knowledge of the spectral characteristics of clean speech and is used to quantise features from noise-corrupted speech. Our comparative results from the ETSI Aurora-2 recognition task show that the perceptually-weighted vector quantisation of LFBEs achieves higher recognition accuracies for noisy speech than the unweighted vector quantisation, memoryless and multi-frame GMM-based block quantisation and scalar quantisation of Mel frequency-warped cepstral coefficients.

Journal Title
Conference Title

9th European Conference on Speech Communication and Technology

Book Title
Edition
Volume
Issue
Thesis Type
Degree Program
School
Publisher link
DOI
Patent number
Funder(s)
Grant identifier(s)
Rights Statement
Rights Statement
Item Access Status
Note
Access the data
Related item(s)
Subject
Persistent link to this record
Citation