Creativeguy Persian ASR: Speech Recognition Dataset with 18808 Samples

18808 audio samples of Persian speech, totaling 143 hours and 45 minutes of audio. The dataset is split into 13165 training samples and 5643 test samples. It was uploaded by user 'veziriii' to Hugging Face in May 2026.

Use Cases

• Train Persian speech-to-text models based on the provided audio and transcript pairs.
• Benchmark ASR model performance on Persian language using the defined test split.
• Fine-tune multilingual speech models on Persian audio data.

Strengths

• Contains 18808 audio samples, providing a substantial volume of data.
• Offers a clear split of 13165 training and 5643 test samples for model development and evaluation.
• Total audio duration is 143 hours and 45 minutes, a significant amount for 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.
• Last updated 2026-05-26 08:22:45; freshness should be verified.

Provenance

Source

huggingface

Freshness

Last updated 2026-05-26 08:22:45.