search

CMsmartvoice / One-Shot-Voice-Cloning

:relaxed: One Shot Voice Cloning base on Unet-TTS
236 stars 41 forks source link

what is this #1

Open hdmjdp opened 2 years ago

hdmjdp commented 2 years ago

https://github.com/CMsmartvoice/One-Shot-Voice-Cloning/blob/091bfefaa34427abad0653621d412a1886a71a58/TensorFlowTTS/tensorflow_tts/models/unetts.py#L69

sheng? yun?

CMsmartvoice commented 2 years ago

The Chinese language is composed of syllables, all of which can be modeled with two phonemes, generally called "vowels" and "rhymes". Vowels can be understood as consonants and rhymes as vowels. Based on this, I calculate statistics for the duration information of the two types of phonemes separately.