Adaptation to New Microphones Using Tied-Mixture Normalization

Abstract

In this paper, we present several approaches designed to increase the robustness of BYBLOS, the BBN continuous speech recognition system. We address the problem of increased degradation in performance when there is mismatch in the characteristics of the training and the test microphones. We introduce a new supervised adaptation algorithm that computes a transformation from the training microphone codebook to that of a new microphone, given some information about the new microphone. Results are reported for the development and evaluation test sets of the 1993 ARPA CSR Spoke 6 WSJ task, which consist of speech recorded with two alternate microphones, a stand-mount and a telephone microphone.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1994
Accession Number
ADA459585

Entities

People

  • Anastasios Anastasakos
  • Francis Kubala
  • John Makhoul
  • Richard Schwartz

Organizations

  • BBN Technologies

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Automated Speech Recognition
  • Covariance
  • Degradation
  • Errors
  • Gaussian Distributions
  • Hidden Markov Models
  • Language
  • Markov Models
  • Microphones
  • Models
  • Natural Language Processing
  • Probability
  • Signal Processing
  • Standards
  • Test Sets
  • Training

Fields of Study

  • Computer science

Readers

  • Graph Algorithms and Convex Optimization.
  • Speech Processing/Speech Recognition.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks