nagohachi/

reazonspeech_qwen3-asr_large

This is the ReazonSpeech corpus's large split, featuring Qwen3-ASR 1.7B transcriptions. Since the original transcriptions often contain errors, comparing them with the Qwen3-ASR outputs could be useful.

import json import webdataset as wds from huggingface_hub import get_token

SHARDS = ( f"pipe:curl -sLf -H 'Authorization: Bearer {get_token()}' " "https://huggingface.co/datasets/nagohachi/reazonspeech_qwen3-asr_large/"… See the full description on the dataset page: https://huggingface.co/datasets/nagohachi/reazonspeech_qwen3-asr_large.