Related Experiment Videos
Control of voice F0 by an artificial neural network
1Boys Town National Research Hospital, Omaha, Nebraska 68131.
The Journal of the Acoustical Society of America
|September 1, 1994
Summary
A new artificial neural network model demonstrates effective fundamental frequency (F0) control using only seven neurodes. This brain-inspired model offers insights into the neural mechanisms of vocalization and speech production.
Area of Science:
- Computational Neuroscience
- Speech Science
- Artificial Intelligence
Background:
- Understanding the neural basis of vocal control is crucial for speech production research.
- Existing models may not fully capture the complexity of fundamental frequency (F0) regulation.
- Artificial neural networks offer a powerful tool for modeling biological systems.
Purpose of the Study:
- To present a novel artificial neural network model for simulating fundamental frequency (F0) control.
- To investigate the minimal neural components required for effective F0 regulation.
- To explore the potential of this model in understanding brain mechanisms of vocalization.
Main Methods:
- Development of an artificial neural network with seven neurodes.
- Simulation of neural connections including motor, inhibitory, and excitatory neurodes.
- Analysis of the network's capacity for achieving precise F0 control.
Main Results:
- The model achieved good fundamental frequency (F0) control with a compact network of seven neurodes.
- Specific configurations of inhibitory and excitatory neurodes were identified as key for control.
- The model demonstrated the feasibility of simplified neural architectures for complex vocal functions.
Conclusions:
- The proposed artificial neural network model provides a plausible mechanism for a component of brain-based F0 control.
- This simplified model highlights the efficiency of neural computation in vocalization.
- The model serves as a valuable tool for future research into the neurobiology of speech.