Here's a refined and paraphrased version of your dataset description to make it clearer and more efficient:
Total Audio Files: 673,659
Total Audio Length: 856.20 hours
Total Audio Collected Data: 18 OCT, 2024
Department-wise Statistics (After Removal)
STT_PC
73,894
78.90… See the full description on the dataset page:
https://huggingface.co/datasets/openpecha/TTS_training_data.