Synthetic Egyptian Arabic Speech Dataset with Quality Metadata

A curated collection of 1000 Egyptian Arabic speech samples, each representing the best audio selected from four different text-to-speech models. The dataset was created by MohamedGomaa30 and includes transcription text and quality metadata for each entry. It was last updated on May 14, 2026.

Use Cases

• Benchmarking text-to-speech models based on the quality metadata for each audio sample.
• Training or fine-tuning speech recognition systems for Egyptian Arabic based on the provided transcriptions.
• Studying the characteristics and quality of synthetic speech from different TTS architectures.
• Developing tools for automatic speech quality assessment using the curated best-of-four selections.

Strengths

• Contains 1000 distinct speech samples, providing a substantial base for analysis.
• Each sample is the best audio selected from four TTS models, suggesting a quality-filtered collection.
• Includes metadata indicating which of the four models produced the selected audio, allowing for model comparison.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

huggingface

Collection Method

Synthetically generated by selecting the best output from four text-to-speech models for each prompt.

Freshness

Last updated 2026-05-14 14:22:33; freshness should be verified.

Geography

Egyptian Arabic dialect.