The Persian Speech Recognition Benchmark (PSRB) is a comprehensive dataset designed to evaluate Persian Automatic Speech Recognition (ASR) systems under diverse real-world conditions.This 1-hour sample provides a representative subset of the full PSRB corpus, capturing various accents, speech styles, speaker demographics, and acoustic environments.
Supported Tasks and… See the full description on the dataset page: https://huggingface.co/datasets/PartAI/PSRB.
