Closed orenstuf closed 4 months ago
Hi @orenstuf,
The LJ Speech checkpoint is only trained on LJ Speech corpus. Being a read speech corpus it doesn't really have a lot of variation or interrogative sentences for the model to learn.
The only mitigation for this will be to improve the dataset. Or try fine-tuning the model to your voice. Record 20 such sentences and fine-tune and see if the model learns something from it.
Regards, Shivam
Hello,
Here are some issues with interrogative sentences:
"May I take your order?" - sounds like a standard declarative sentence.
"Where do you live?" - sounds like a standard declarative sentence and adds stress on the word "you".
"Did you finish your homework on time?" - sounds like a question, but the intonation in incorrect.
Can you give me some guidelines on how to train matcha to overcome these issues?
Thanks a lot