The OCON model: an old but gold solution for distributable supervised classification

TL;DR

This paper presents the OCON model, a simple yet effective approach for supervised classification in speech recognition, achieving high accuracy with better generalization and distributed applicability through systematic tuning and search methods.

Contribution

It introduces the OCON model with a structured One-Class approach, demonstrating competitive accuracy and enhanced generalization in speech phoneme classification.

Findings

Achieved 90.0 - 93.7% classification accuracy.

Model generalizes well across language contexts.

Open-source code available on GitHub.

Abstract

This paper introduces to a structured application of the One-Class approach and the One-Class-One-Network model for supervised classification tasks, specifically addressing a vowel phonemes classification case study within the Automatic Speech Recognition research field. Through pseudo-Neural Architecture Search and Hyper-Parameters Tuning experiments conducted with an informed grid-search methodology, we achieve classification accuracy comparable to nowadays complex architectures (90.0 - 93.7%). Despite its simplicity, our model prioritizes generalization of language context and distributed applicability, supported by relevant statistical and performance metrics. The experiments code is openly available at our GitHub.