Speech signal processing system and method for Slovenian-specific transcription and speech synthesis training

Abstract

A system and method for processing speech signals for transcription, specifically for the Slovenian language, and learning speech synthesis solves the above-mentioned technical problem by providing a computer-implemented method and system configured to process Slovenian speech sound samples obtained from multiple speakers. The system extracts phonetic and prosodic features from the sound signals and generates normalized transcription data that reduces inter-speaker variability while preserving the prosodic structure specific to the Slovenian language.