Skip to content

Synthesized speech output limited to 0-2 seconds - Possible preprocessing issue? #17

@dahyunnss

Description

@dahyunnss

I'm experiencing an issue with the synthesized speech generated by the StyleSpeech model.

All of the synthesized audio clips are only 0-2 seconds long, regardless of the input text length.
I suspect there might be a problem in the preprocessing stage. Could this be causing the synthesized speech to be cut short?
Has anyone else encountered this issue? If so, how did you resolve it?
I'd appreciate any insights or suggestions on how to troubleshoot this problem.
Thank you for your help!"
(I used the same TextGrid files you provided and preprocessed everything into npy files for my work.)

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions