Congratulations on the acceptance of both SingAligner and FT-GAN! Specifically, I'd like to inquire about the alignment efficacy of your SingAligner on opera singing voice. In your paper "FT-GAN: Fine-Grained Tune Modeling for Chinese Opera Synthesis," why did you choose to utilize traditional MFA alignment tools instead of your proposed SingAligner for Alignment Annotation? Was this decision based on the superior reliability of MFA's performance?
Congratulations on the acceptance of both SingAligner and FT-GAN! Specifically, I'd like to inquire about the alignment efficacy of your SingAligner on opera singing voice. In your paper "FT-GAN: Fine-Grained Tune Modeling for Chinese Opera Synthesis," why did you choose to utilize traditional MFA alignment tools instead of your proposed SingAligner for Alignment Annotation? Was this decision based on the superior reliability of MFA's performance?