More Information needed
Combined word-aligned dataset for training Akan/Twi TTS models. Audio encoded with WavTokenizer (75H