ZLSCompLing/LOD_Claude

LOD_Claude Dataset Dataset Description LOD_Claude is a Luxembourgish speech dataset containing audio recordings paired with transcriptions. The audio features a synthetic voice named Claude reading example sentences from the LOD (Lëtzebuerger Online Dictionnaire) available at lod.lu. Dataset Statistics Total samples: 39,034 Training samples: 37,084 Validation samples: 1,950 Language: Luxembourgish (Lëtzebuergesch) Audio format: WAV files Sample… See the full description on the dataset page: https://huggingface.co/datasets/ZLSCompLing/LOD_Claude.