Augmented intelligence with voice assistance and automated machine learning in Industry 5.0

Area of Science:

• Artificial Intelligence
• Human-Computer Interaction
• Industrial Engineering

Background:

• Existing voice assistants lack data-driven algorithms, relying on knowledge management and simulations.
• Practical applications and real-life evaluations of voice assistants in industrial contexts are limited.
• Augmented intelligence requires socio-technological systems where human and artificial agents co-evolve.

Purpose of the Study:

• To propose and evaluate the integration of voice assistance technology with Automated Machine Learning (AutoML).
• To enable the augmented intelligence paradigm within the Industry 5.0 context.
• To facilitate intuitive, voice-based interaction with machine learning pipelines for immediate task-related insights.

Main Methods:

• Integration of Speech-To-Text (STT) and Text-To-Speech (TTS) technologies with AutoML.
• Development of voice-controlled interaction with automatically generated Machine Learning (ML) pipelines.
• Evaluation of the proposed approach in a real-world manufacturing environment using a structured methodology.

Main Results:

• The proposed voice-assisted AutoML approach demonstrated effectiveness in a practical manufacturing setting.
• Real-time voice interaction with ML pipelines provided immediate, actionable insights to users.
• The system facilitated a more intuitive and efficient human-AI collaboration.

Conclusions:

• The integration of voice assistance and AutoML is a viable strategy for realizing augmented intelligence in Industry 5.0.
• This approach enhances human-AI collaboration by enabling seamless voice interaction with ML processes.
• The findings support the effectiveness of data-driven, voice-enabled solutions in industrial applications.