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 Videos

Articulatory representation and speech technology

O Schmidbauer1, F Casacuberta, M J Castro

  • 1Siemens AG Munich, FRG.

Language and Speech
|April 1, 1993
PubMed
Summary
This summary is machine-generated.

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

Impact of universal use of the McGrath videolaryngoscope as the first option for all intubations in the operating room: The multicentre prospective before-after VIDEOLAR-SURGERY study protocol.

Revista espanola de anestesiologia y reanimacion..·2024
Same author

Global phylogenomic assessment of Leptoseris and Agaricia reveals substantial undescribed diversity at mesophotic depths.

BMC biology·2023
Same author

Ion Transport Mechanisms in Sea Urchin Sperm.

Development, growth & differentiation·2023
Same author

Deep Learning-Based Feature Extraction of Acoustic Emission Signals for Monitoring Wear of Grinding Wheels.

Sensors (Basel, Switzerland)·2022
Same author

Probabilistic tsunami forecasting for early warning.

Nature communications·2021
Same author

An Assessment of Earth's Climate Sensitivity Using Multiple Lines of Evidence.

Reviews of geophysics (Washington, D.C. : 1985)·2020
Same journal

Does the Majority Language Shape the Pragmatic Relevance of Prosody? Evidence From Heritage Speakers of Turkish in Germany.

Language and speech·2026
Same journal

Frequency Effects in the L2 Acquisition of French Liaison: A Corpus-Based Adaptation of Gradient Symbolic Representations.

Language and speech·2026
Same journal

Prominence and Grouping in Papuan Malay Prosody Perception.

Language and speech·2026
Same journal

Perceptual Tuning to Structure: Integrating the Phonetic Detail of Coarticulatory Vowel Nasalization With Prosodic and Information Structure.

Language and speech·2026
Same journal

An Investigation of the Phonetic Variation of the Word-Initial /l/ and /n/ Across Regional Varieties of Mandarin.

Language and speech·2026
Same journal

How Much Do We Imitate in a Non-Native Language? Explicit Versus Implicit Imitation in L2 Speech.

Language and speech·2026
See all related articles

This study shows articulation-based models are effective for speech recognition and synthesis. These methods improve speaker-independent recognition and can enhance synthesized speech quality.

Area of Science:

  • Speech Technology
  • Acoustic Phonetics
  • Computational Linguistics

Background:

  • Traditional speech recognition and synthesis models often lack detailed articulatory information.
  • Bridging the gap between acoustic signals and articulatory movements is crucial for advancing speech technology.

Purpose of the Study:

  • To demonstrate the feasibility and utility of articulation-based approaches in speech recognition and synthesis.
  • To develop and evaluate articulatory models for improved speech technology applications.

Main Methods:

  • Developed an articulatory recognition model estimating manner and place of articulation probabilities.
  • Utilized hidden Markov models and multi-layer perceptrons for articulatory-to-symbolic transformation.
  • Employed an analysis-by-synthesis model to map acoustic signals to articulatory parameters, using EMA measurements.

Related Experiment Videos

  • Investigated a physical model for glottal source generation via numerical simulation for speech synthesis.

    • Main Results:

    • Articulatory approach proves effective for speaker-independent and speaker-adaptive speech recognition.
    • EMA measurements provide valuable initial estimates for articulatory parameters.
    • Numerical simulations of glottal source generation offer insights for improving articulatory synthesis models.

    Conclusions:

    • Articulation-based methods offer a robust foundation for advanced speech recognition systems.
    • Further development of realistic articulatory models is promising for both recognition and synthesis.
    • Integration of physical models and articulatory data enhances the accuracy and quality of speech technology.