Harf-Speech: A Clinically Aligned Framework for Arabic Phoneme-Level Speech Assessment

2026-03-11

arXiv.org

AI Summary

This study addresses the lack of clinically validated tools for automatic phoneme-level pronunciation assessment in Arabic, which hinders scalable speech therapy and language learning. The authors propose the first interpretable, modular evaluation framework that integrates clinical scoring criteria, leveraging Modern Standard Arabic (MSA) phonemic transcriptions, a fine-tuned ASR model (OmniASR-CTC-1B-v2), Levenshtein alignment, and a hybrid scoring mechanism based on longest common subsequence and edit distance. Evaluated on 40 clinical speech samples, the system achieves a Pearson correlation coefficient of 0.791 and an ICC(2,1) of 0.659 with expert ratings—significantly outperforming existing end-to-end approaches and approaching inter-rater reliability levels among clinicians.

Abstract

Automated phoneme-level pronunciation assessment is vital for scalable speech therapy and language learning, yet validated tools for Arabic remain scarce. We present Harf-Speech, a modular system scoring Arabic pronunciation at the phoneme level on a clinical scale. It combines an MSA phonetizer, a fine-tuned speech-to-phoneme model, Levenshtein alignment, and a blended scorer using longest common subsequence and edit-distance metrics. We fine-tune three ASR architectures on Arabic phoneme data and benchmark them with zero-shot multimodal models; the best, OmniASR-CTC-1B-v2, achieves 8.92\% phoneme error rate. Three certified speech-language pathologists independently scored 40 utterances for clinical validation. Harf-Speech attains a Pearson correlation of 0.791 and ICC(2,1) of 0.659 with mean expert scores, outperforming existing end-to-end assessment frameworks. These results show Harf-Speech yields clinically aligned, interpretable scores comparable to inter-rater expert agreement.

Problem

• Arabic
• phoneme-level assessment
• speech therapy
• pronunciation evaluation
• clinical validation

Innovation

• phoneme-level assessment
• Arabic speech evaluation
• clinically aligned scoring
• fine-tuned ASR
• Levenshtein alignment