Moroccan Darija TTS Dataset
This dataset contains Moroccan Darija speech recordings and their corresponding transcriptions, intended for fine-tuning text-to-speech models.
Dataset Structure
wavs_16k/: Directory containing 16kHz mono 16-bit WAV audio files.metadata_train.csv: CSV file with training data.metadata_val.csv: CSV file with validation data.
Usage
Load the dataset using:
from datasets import load_dataset
dataset = load_dataset("Fahd1199/darija-tts")