RoxasYTB/defective_turret_de-ljspeech

Defective Turret — Deutsch (de) LJSpeech dataset of Defective Turret (de). 175 pairs 44.1kHz mono 16-bit PCM WAV (original wiki quality) Piper TTS Training (High Quality on T4 GPU) Preprocessing (downsample to 22.05kHz) python3 -m piper_train.preprocess \ --language de \ --input-dir ./defective_turret_de \ --output-dir ./train_defective_turret_de \ --dataset-format ljspeech \ --single-speaker \ --sample-rate 22050 Training… See the full description on the dataset page: https://huggingface.co/datasets/RoxasYTB/defective_turret_de-ljspeech.