You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In the code, group MFA inputs for better parallelism. For multi speaker, it maybe go wrong.
For input g_uang3 zh_ou1 n_v3 d_a4 x_ve2 sh_eng1 d_eng1 sh_an1 sh_i1 l_ian2 s_i4 t_ian1 j_ing3 f_ang1 zh_ao3 d_ao4 i2 s_i4 n_v3 sh_i1.
The TexGrid is
The mfa code is borrowed from the official repo of NATSpeech. Unfortunately I have not studied the mfa module in depth. We encourage you to push your commits on multi-speaker chinese corpus. Thanks a lot!
hello, I push a commit. But I don't know it suits for your planning. The parameter of mfa_group seems unnecessary. It doesn't prompt the train speed compared with wetts(it uses version 2.1.0). Also need add the parameter of spk_name.
In the code, group MFA inputs for better parallelism. For multi speaker, it maybe go wrong.
For input
g_uang3 zh_ou1 n_v3 d_a4 x_ve2 sh_eng1 d_eng1 sh_an1 sh_i1 l_ian2 s_i4 t_ian1 j_ing3 f_ang1 zh_ao3 d_ao4 i2 s_i4 n_v3 sh_i1
.The TexGrid is
The text was updated successfully, but these errors were encountered: