search

TensorSpeech / TensorFlowTTS

:stuck_out_tongue_closed_eyes: TensorFlowTTS: Real-Time State-of-the-art Speech Synthesis for Tensorflow 2 (supported including English, French, Korean, Chinese, German and Easy to adapt for other languages)
https://tensorspeech.github.io/TensorFlowTTS/
Apache License 2.0
3.85k stars 815 forks source link

Trying to run mfa for Korean Fastspeech2, does anyone know where I can find Korean lexicon? #385

Closed charlie-codes-choi closed 3 years ago

charlie-codes-choi commented 4 years ago

EoM

ZDisket commented 4 years ago

@charlie-codes-choi You can find a pretrained acoustic and G2P Korean model here: https://montreal-forced-aligner.readthedocs.io/en/latest/pretrained_models.html Since there are no dictionaries, you'll have to generate one. The documentation shows how to: https://montreal-forced-aligner.readthedocs.io/en/latest/g2p_dictionary_generating.html

charlie-codes-choi commented 4 years ago

Fantastic. Thanks a lot ZDisket! I will try it and share how it went here.

Jackson-Kang commented 4 years ago

Hello, @ZDisket and @charlie-codes-choi .

The pretrained Korean MFA model has an issue. Please see the below. https://github.com/HGU-DLLAB/Korean-FastSpeech2-Pytorch/issues/3#issuecomment-731979268

I recommend to generate dictionary for your own dataset using g2pk module (https://github.com/Kyubyong/g2pK).

stale[bot] commented 3 years ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.