Habibi: Laying the Open-Source Foundation of Unified-Dialectal Arabic Speech Synthesis

Abstract  Arabic spans over 30 spoken varieties, yet no open-source text-to-speech system unifies them. Key barriers include substantial cross-dialect lexical and phonological divergence, scarce synthesis-grade data, and the absence of a standardized multi-dialect evaluation benchmark. We present Habibi, a unified-dialectal Arabic TTS framework that addresses all three. Through a multi-step curation pipeline, we repurpose open-source ASR corpora into TTS training data covering 12+ regional dialects. A linguistically-informed curriculum learning strategy—progressing from Modern Standard Arabic to dialectal data—enables robust zero-shot synthesis without text diacritization. We further release the first standardized multi-dialect Arabic TTS benchmark, comprising over 11,000 utterances across 7 dialect subsets with manually verified transcripts. On this benchmark, our unified model matches or surpasses per-dialect specialized models. Both automatic metrics and human evaluations confirm that Habibi is highly competitive with ElevenLabs' Eleven v3 (alpha) in intelligibility, speaker similarity, and naturalness. Extensive ablations (~8,000 H100 GPU hours, 30+ configurations) validate each design choice. We open-source all checkpoints, training and inference code, and benchmark data—the first such release for multi-dialect Arabic TTS.

Figure 1  Our open-source unified-dialectal model (Habibi-Unified) vs. ElevenLabs' commercial Eleven v3 (alpha).
WER-S/O: word error rates from two different ASR systems. D/S/N-MOS: dialect pronunciation accuracy, speaker similarity, and naturalness.

Contents

• Zero-Shot Speech Synthesis Performance
• Code-Switching Performance

This page is for research demonstration purposes only.

Zero-Shot Speech Synthesis Performance

Code-Switching Performance