Sound Source Separation Model Tuning With Neural Architecture Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sound source separation deep learning models, such as Conv-TasNet, require significant time and computing resources for redesign and hyperparameter modification across various domains, necessitating expert intervention and limiting their efficiency and performance.
Innovation Solution
A neural architecture search (NAS) algorithm is used to automatically search for and reconstruct hyperparameter combinations in sound source separation deep learning models, optimizing them for specific acoustic data applications while minimizing human intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual redesign and hyperparameter modification are performed to apply the model to various domains, then performance can be optimized for specific domains, but it requires significant time and computer resources along with expert intervention
Solution Approach 1:
The patent applies self-service by enabling the system to automatically optimize its own architecture and hyperparameters through neural architecture search (NAS) algorithms. The model autonomously performs redesign and parameter modification without requiring manual expert intervention, thereby reducing time loss while maintaining optimized performance for various domains
Solution Approach 2:
The patent utilizes parameter changes by systematically modifying hyperparameters and architectural parameters through automated search algorithms. The NAS process explores different parameter configurations (kernel sizes, channel dimensions, layer depths) to automatically identify optimal settings for each domain, resolving the contradiction between performance optimization and time consumption
2Reliability
If manual redesign and hyperparameter modification are performed to apply the model to various domains, then performance can be optimized for specific domains, but it requires significant computer resources
Solution Approach 1:
The system performs self-service by automatically conducting neural architecture search and hyperparameter optimization using computational algorithms. This automated process eliminates the need for manual expert intervention and reduces the computational resources required compared to traditional manual tuning methods, while still achieving domain-specific performance optimization
Solution Approach 2:
The patent replaces the mechanical system of manual expert intervention with an automated computational system. The neural architecture search algorithm substitutes human expertise with algorithmic optimization, reducing computing resource consumption while maintaining the ability to optimize model performance across various domains
3Reliability
If manual redesign and hyperparameter modification are performed to apply the model to various domains, then performance can be optimized for specific domains, but it can be performed by only deep learning experts
Solution Approach 1:
The patent implements self-service by enabling the system to automatically perform architecture redesign and hyperparameter optimization through integrated NAS algorithms. This eliminates the requirement for deep learning experts to manually tune parameters, making the process accessible to users without specialized knowledge while maintaining optimized performance
Solution Approach 2:
The patent introduces an intermediary system in the form of automated neural architecture search algorithms. This intermediary layer between the user and the model optimization process handles the complex tasks of architecture redesign and parameter tuning, simplifying operation for users while ensuring optimal performance through algorithmic optimization
Data Source
AI summary
Disclosed are a system and method for automating the design of a sound source separation deep learning model. A method of automating a design of a sound source separation deep learning model, which is performed by a design automation system, may include automatically searching for a combination of hyper parameters of a separation model constructed in a sound source separation deep learning model by using a neural architecture search (NAS) algorithm and reconstructing the sound source separation deep learning model based on the retrieved combination of the hyper parameters of the separation model.