aiqbench analysis
This paper advances speech synthesis research by extending the DYNARTmo dynamic articulatory model with detailed acoustic and aerodynamic simulations to better mimic human vocal production. The work focuses on optimizing the model for sound recognizability to serve as a controllable plant for future neural simulations of early language acquisition and babbling. While it involves audio generation, the core contribution lies in the underlying machine learning and computational linguistics frameworks rather than consumer audio hardware or general software updates.
Vocal-Tract Acoustics, Aerodynamics, and Overlap-and-Add Synthesis in DYNARTmo
arXiv:2610.24490v1 Announce Type: new Abstract: This paper presents the acoustic and aerodynamic extension of DYNARTmo, a dynamic articulatory model for visualizing and synthesizing speech production. Gestural control trajectories determine vocal-tract geometry, which is converted into an acoustic tube with nasal branching and effective lateral passages. A lumped aerodynamic network relates prescribed glottal motion and pulmonary pressure to airflow and supraglottal pressure storage. Phonatory excitation, oral friction, and glottal aspiration are synthesized as separate contributions. Frame-specific acoustic responses are combined by overlap-and-add, preserving response tails beyond individual control frames. Three German words, ``Tasse'', ``Muse'', and ``Lade'', illustrate the coordination of articulation, pressure, flow, and acoustic sources across vowels, nasals, laterals, fricatives, and plosives. The implementation has been optimized by the author primarily for sound recognizability; independent perceptual evaluation remains outstanding, and fricative-noise naturalness requires further refinement. The model provides an inspectable, articulatorily controlled production plant for future neural simulations of babbling, imitation, and early word acquisition. Further work will improve acoustic realism, reduce empirical source corrections, and extend the sensory information available to developmental learning models.
Abstract:This paper presents the acoustic and aerodynamic extension of DYNARTmo, a dynamic articulatory model for visualizing and synthesizing speech production. Gestural control trajectories determine vocal-tract geometry, which is converted into an acoustic tube with nasal branching and effective lateral passages. A lumped aerodynamic network relates prescribed glottal motion and pulmonary pressure to airflow and supraglottal pressure storage. Phonatory excitation, oral friction, and glottal aspiration are synthesized as separate contributions. Frame-specific acoustic responses are combined by overlap-and-add, preserving response tails beyond individual control frames. Three German words, ``Tasse'', ``Muse'', and ``Lade'', illustrate the coordination of articulation, pressure, flow, and acoustic sources across vowels, nasals, laterals, fricatives, and plosives. The implementation has been optimized by the author primarily for sound recognizability; independent perceptual evaluation remains outstanding, and fricative-noise naturalness requires further refinement. The model provides an inspectable, articulatorily controlled production plant for future neural simulations of babbling, imitation, and early word acquisition. Further work will improve acoustic realism, reduce empirical source corrections, and extend the sensory information available to developmental learning models.
6 figures; ancillary Python reference implementation and word-transcription guide. No audio recordings or saved timelines
Computation and Language (cs.CL)
arXiv:2610.24490 [cs.CL]
(or arXiv:2610.24490v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2610.24490 Focus to learn more arXiv-issued DOI via DataCite
LICENSE_SCOPE.txt
reference_core/CITATION.cff
reference_core/MANIFEST.in
reference_core/MODEL.md
reference_core/README.md
reference_core/VALIDATION.md
reference_core/WORD_EXAMPLES.md
reference_core/examples/single_geometry.py
reference_core/examples/word_transcriptions.txt
reference_core/pyproject.toml
reference_core/src/speech_reference/__init__.py
reference_core/src/speech_reference/aerodynamics.py
reference_core/src/speech_reference/validation.py
reference_core/src/speech_reference/waveguide.py
reference_core/tests/test_reference.py
reference_core/validation/baseline.json
reference_core/validation/cases.json
reference_core/validation/dart_reference.json
reference_core/validation/report.json
(17 additional files not shown) You must enabled JavaScript to view entire file list.
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 311 words