Physics-Informed Neural Engine Sound Modeling with Differentiable Pulse-Train Synthesis

Dimensions overall score 9.0

GitHub Code Pulse

No public code linked for this paper yet.

• The Pulse-Train-Resonator (PTR) model is a differentiable synthesis architecture for generating engine audio.technical 95%
• PTR integrates physics-informed inductive biases, including harmonic decay and valve-dynamics envelopes.technical 90%
• PTR achieves a 21% improvement in harmonic reconstruction compared to a harmonic-plus-noise baseline model.result 90%
• PTR achieves a 5.7% reduction in total loss compared to a harmonic-plus-noise baseline model.result 90%
• The PTR model was validated on 7.5 hours of audio from three diverse engine types.method 95%
• PTR's reliance on physics-informed biases may limit its generalizability beyond engine sounds.limitation 80%
• The need for specialized knowledge in exhaust and engine dynamics might inhibit wide adoption of PTR.limitation 75%
• The PTR model provides interpretable parameters corresponding to physical phenomena.technical 85%
• The Pulse-Train-Resonator (PTR) model directly models engine sounds as parameterized pulse trains propagated through recursive Karplus-Strong resonators.method 70%
• The PTR model integrates physics-informed inductive biases including harmonic decay, thermodynamic pitch modulation, valve-dynamics envelopes, and exhaust system resonances.method 70%
• The PTR model achieves a 21% improvement in harmonic reconstruction compared to a harmonic-plus-noise baseline model.result 70%
• The PTR model achieves a 5.7% reduction in total loss compared to a harmonic-plus-noise baseline model.result 70%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card