Cleaned TTS dataset for Tagin (tgj), a North East Indian language. Derived from the Vaani dataset with SNR filtering, LUFS normalization, and text cleaning.

	Stats

Metric Value

Total clips 80

Total hours 0.1h

High SNR (>=20dB) 37 clips

Medium SNR (15-20dB) 43 clips

Sample rate 16kHz

Audio format WAV, 16-bit PCM

	Schema

Column Type Description

audio Audio 16kHz WAV audio

text string Cleaned… See the full description on the dataset page: https://huggingface.co/datasets/sulabhkatiyar/ne-tts-tgj.