Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jul 7, 2026

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example
08:45

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Published on: October 24, 2012

Acoustic-to-phonetic mapping using recurrent neural networks.

M D Hanes1, S C Ahalt, A K Krishnamurthy

  • 1Dept. of Electr. Eng., Ohio State Univ., Columbus, OH.

IEEE Transactions on Neural Networks
|January 1, 1994
PubMed
Summary

Artificial neural networks effectively map acoustic speech signals to phonetic representations. Simple Elman recurrent networks accurately identify consonants and vowels across varying speech tempos.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Optical implementation and applications of closest-vector selection in neural networks.

Applied optics·2010
Same author

Comparison between single-step and balloon dilatational tracheostomy.

British journal of anaesthesia·2010
Same author

Diffusion approximation of frequency sensitive competitive learning.

IEEE transactions on neural networks·1997
Same author

Codeword distribution for frequency sensitive competitive learning with one-dimensional input data.

IEEE transactions on neural networks·1996
Same author

Intensity-weighted average of instantaneous frequency as a model for frequency discrimination.

The Journal of the Acoustical Society of America·1993
Same author

A critical review of electroglottography.

Critical reviews in biomedical engineering·1985

Area of Science:

  • Speech Recognition
  • Artificial Intelligence
  • Acoustic Phonetics

Background:

  • Speech recognition systems require accurate acoustic-to-phonetic mapping.
  • Temporal dynamics of speech signals are crucial for accurate phonetic interpretation.
  • Previous methods often struggle with variable speech rates.

Purpose of the Study:

  • To apply artificial neural networks for acoustic-to-phonetic mapping.
  • To evaluate the performance of Elman simple recurrent networks on this task.
  • To assess the networks' ability to handle different speech tempos.

Main Methods:

  • Utilized Elman simple recurrent neural networks.
  • Trained networks using standard backpropagation techniques.
  • Experimented with formant contour data from speech spoken at slow and normal tempos.

More Related Videos

Neuronavigation-guided Repetitive Transcranial Magnetic Stimulation for Aphasia
08:48

Neuronavigation-guided Repetitive Transcranial Magnetic Stimulation for Aphasia

Published on: May 6, 2016

Related Experiment Videos

Last Updated: Jul 7, 2026

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example
08:45

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Published on: October 24, 2012

Neuronavigation-guided Repetitive Transcranial Magnetic Stimulation for Aphasia
08:48

Neuronavigation-guided Repetitive Transcranial Magnetic Stimulation for Aphasia

Published on: May 6, 2016

Main Results:

  • Elman networks successfully performed acoustic-to-phonetic mapping.
  • Networks accurately mapped formant contours to CVC syllables.
  • Demonstrated capability in consonant discrimination and vowel identification.

Conclusions:

  • Simple recurrent neural networks are effective for acoustic-to-phonetic tasks.
  • The proposed method is robust to variations in speech rate.
  • This approach offers a viable solution for real-time speech recognition.