Corpus ID: 15812895

Auditory Effects for ASR

  title={Auditory Effects for ASR},
  author={R. Lyon},
  • R. Lyon
  • Published 1996
  • Almost all ASR front ends use an amplitude-independent representation of spectral shape as the primary feature vector, obtained via some combination of normalization, logarithms, or AR modeling. They also typically represent total power or loudness as a separate feature. These ideas are fine to first order, and have gotten ASR to where it is today. But they totally punt on the issue of what is "loud enough". 
    2 Citations
    Polyglot Speech Synthesis: A Review
    • 3
    Polyglot Speech Synthesis Based on Cross-Lingual Frame Selection Using Auditory and Articulatory Features
    • 12


    Self-normalization and noise-robustness in early auditory representations
    • 122
    On the importance of time—a temporal representation of sound
    • 156
    • PDF
    A computational model of binaural localization and separation
    • 152
    • PDF
    Auditory representations of acoustic signals
    • 383
    • PDF
    A computational model of filtering, detection, and compression in the cochlea
    • 396
    • PDF
    The GRASP sound separation system
    • 28
    Correlograms and the Separation of Sounds
    • R. Duda, R. Lyon, M. Slaney
    • Computer Science
    • 1990 Conference Record Twenty-Fourth Asilomar Conference on Signals, Systems and Computers, 1990.
    • 1990
    • 43
    • PDF
    A theory and computational model of auditory monaural sound separation
    • 140
    Computational models of neural auditory processing
    • 93
    • PDF
    Experiments in isolated digit recognition with a cochlear model
    • E. Loeb, R. Lyon
    • Computer Science
    • ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing
    • 1987
    • 10
    • PDF