Persian Farsi Speech Dataset for Text-to-Speech Model Training

A Persian (Farsi) text-to-speech corpus built by concatenating and denoising multiple existing Farsi datasets. The dataset was created by author Thomcles and last updated on November 9, 2025. It is intended for training and evaluation of speech synthesis models in Persian.

Use Cases

• Train text-to-speech models based on the described Persian audio corpus.
• Evaluate speech synthesis model performance based on the described clean audio benchmark.
• Fine-tune pre-trained TTS models for Persian based on the described dataset.
• Conduct research on Persian speech generation based on the described TTS corpus.

Strengths

• Audio quality was filtered using dnsmos, keeping only clean audio with a mos_ovr score >= 3.0.
• Dataset is specifically built for Persian (Farsi) text-to-speech tasks.

Limitations

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

Provenance

Source

huggingface

Collection Method

Built by concatenating and denoising multiple existing Farsi datasets.

Freshness

Last updated 2025-11-09 08:41:17; freshness should be verified.

Geography

Persian (Farsi) language focus.