Aybee5/hausa-tts

Multi-speaker Hausa text-to-speech dataset. from datasets import load_dataset, Audio # Load the dataset ds = load_dataset("parquet", data_files="hausa_tts_embedded.parquet", split="train") # Cast audio column to Audio type with 24kHz sampling rate ds = ds.cast_column("audio", Audio(sampling_rate=24000)) # Use with Unsloth TTS from transformers import AutoProcessor processor = AutoProcessor.from_pretrained("unsloth/csm-1b")