Thomcles/YodaLingua-Luxembourgish-Extended

YodaLingua-Luxembourgish-Extended YodaLingua is a high-quality speech dataset designed for training text-to-speech (TTS) systems, ASR models, and any application requiring clean, well-aligned audio–text pairs.This release contains the Luxembourgish-Extended portion of the multilingual YodaLingua collection. 🧾 Dataset Overview Property Value Total clips 21,687 audio–transcription pairs Total duration 75 hours Speakers 2,674 distinct speakers Audio… See the full description on the dataset page: https://huggingface.co/datasets/Thomcles/YodaLingua-Luxembourgish-Extended.