JayLL13/vivoice-relabeled
收藏官方服务：
资源简介：
--- dataset_info: features: - name: audio dtype: audio - name: transcript dtype: string - name: source dtype: string --- # Vivoice relabeled using Qwen3-ASR The original dataset: [capleaf/viVoice](https://huggingface.co/datasets/capleaf/viVoice) Sample rate: 44.1 kHz > [!NOTE] > **How-tos**: I use the [Qwen/Qwen3-ASR-1.7B](https://huggingface.co/Qwen/Qwen3-ASR-1.7B) models then update the label for the sample (audio - text) which have WER (word error rate) < 15%.
提供机构：
JayLL13
