The project will transfer the use of language models to industry. The innovative results of the project will allow existing systems across the region to be upgraded and manual input in the systems to be replaced by user-friendly spoken communication. The project will open the way for ŠPICA to enter a new technological niche of optimising connected supply chains and will increase the market for its products.

SPECIFIC OBJECTIVES:

1. Exploring how language and speech technologies can contribute to more effective human-machine communication in industrial environments, taking into account limiting factors such as noise, multilingualism, dialects, etc.
2. Studying and testing technical equipment (microphones, headsets, mobile phones) and technical options for the deployment of language and speech technologies in industrial environments (optimal ways of capturing and transmitting speech given the constraints of industrial environments).
3. Designing an acoustic pre-processor to reduce the presence of noise in captured speech recordings.
4. Developing a multilingual speech recognition model to ensure high recognition accuracy for South Slavic languages and the most spoken Slovenian dialects.
5. Prototype integration of a speech communication system into one of the solutions offered by ŠPICA for process management in production.

RESULTS

• D5.1: Analysis of how speech and language technologies can be used to improve the effectiveness of human-machine communication in industrial environments (August 2024).
• D5.2: Report on the suitability of technical equipment and possible ways to integrate speech and language technologies in production environments (August 2024).
• D5.3: Web service for acoustic signal pre-processing and noise cancellation (August 2025).
• D5.4: An accurate and robust multilingual speech recognition model for South Slavic languages (August 2025).
• D5.5: Prototype of the integration of a speech communication system with a selected business process management solution for manufacturing (June 2026).