Temporal Code Temporal Code Temporal Code (Acoustic Front-end) - - PowerPoint PPT Presentation

temporal code temporal code temporal code
SMART_READER_LITE
LIVE PREVIEW

Temporal Code Temporal Code Temporal Code (Acoustic Front-end) - - PowerPoint PPT Presentation

Temporal Code Temporal Code Temporal Code (Acoustic Front-end) Human Recognition Machine Recognition RECOGNIZED UTTERANCE LANGUAGE MODELING (Back-end) HYPOTHESIZED UTTERANCES STATISTICAL SEQUENCE RECOGNITION ACOUSTIC REPRESENTATION


slide-1
SLIDE 1

Temporal Code

slide-2
SLIDE 2

Temporal Code

slide-3
SLIDE 3

Temporal Code

(Acoustic Front-end)

slide-4
SLIDE 4

Human Recognition

slide-5
SLIDE 5

Machine Recognition

(Back-end)

SPEECH WAVEFORM RECOGNIZED UTTERANCE ACOUSTIC REPRESENTATION HYPOTHESIZED UTTERANCES

ACOUSTIC FRONT END STATISTICAL SEQUENCE RECOGNITION LANGUAGE MODELING

slide-6
SLIDE 6

Human vs. Machine

slide-7
SLIDE 7

“Top-down” Processing

slide-8
SLIDE 8

Machine Training

  • Aurora-4 Speech Database
  • Wall Street Journal (WSJO) Corpus
  • Large Vocabulary Continuous Speech Recognition
  • 7,138 clean speech utterances, 16kHz
slide-9
SLIDE 9

Human Training

  • Wernicke’s Area: Speech Understanding
  • Broca’s Area: Speech Production
slide-10
SLIDE 10

Acoustic Model

Emission Probability Density Transition Probability

Hidden Markov Model (HMM)

  • Each triphone characterized

by HMM consisting of 3 states, 8 Gaussian mixtures per state

slide-11
SLIDE 11

Acoustic Model

  • Maximum likelihood (ML)

training applied to estimate a set of context-dependent triphone acoustic models

slide-12
SLIDE 12

Language Model

  • Standard 5k lexicon

(CMU pronouncing Dictionary)

  • Tri-gram language model
slide-13
SLIDE 13

Decoder

  • Single-pass Viterbi beam

search-based decoder

slide-14
SLIDE 14
  • Noise-Vocoder
  • Tone-Vocoder

Human Recognition

slide-15
SLIDE 15

CI Recognition

slide-16
SLIDE 16

Normal Hearing vs. CI

  • Cochlear Implant range (hatched area) compared

with average normal hearing scores (filled squares)

slide-17
SLIDE 17

CI vs. Machine Recognition

  • ASR provided most accurate simulation ever!
slide-18
SLIDE 18

Machine Recognition

  • ASR derived by world’s best auditory scientists
slide-19
SLIDE 19

Effects of Training

slide-20
SLIDE 20

Effects of Training

slide-21
SLIDE 21

Effects of Training

slide-22
SLIDE 22

Clinical Implications

  • Alter Frequency Allocation
  • Deactivate Interfering Electrodes
  • Alter Compression Curve
  • Modify Electric Pulse Width
slide-23
SLIDE 23

Summary

slide-24
SLIDE 24

Information Technology

  • 2014, HMM can now improve Hearing Science
slide-25
SLIDE 25

Future Work

  • Design improved signal

processing to mimic:

  • 1) Place code of neurons
  • 2) Neural Firing Rates
slide-26
SLIDE 26

FAME Strategy

  • Frequency Amplitude Modulation Encoder
slide-27
SLIDE 27
slide-28
SLIDE 28
slide-29
SLIDE 29

SOUND

Spectral Or Undertone Normalization Decomposition