Closed abanoy closed 5 months ago
I would also like to know for using the api. There's a line for "api_name="/infer_convert"" that states "'F0 curve file (optional). One pitch per line. Replaces the default F0 and pitch modulation:' File component"
I would also like to know for using the api. There's a line for "api_name="/infer_convert"" that states "'F0 curve file (optional). One pitch per line. Replaces the default F0 and pitch modulation:' File component"
That's not needed, it's hided in the page and it's none by default in the following function. It's just a placeholder there. I deleted that component in the page and removed the correspond parameter in the function and passed a None.
This issue was closed because it has been inactive for 15 days since being marked as stale.
This is somewhat an extension of this post in which I don't believe this has been really been explicitly answered to those who want to know the usage of the f0 curve.
I myself am interested in knowing how to use this also. The best that I've attempted was using PRAAT on my audio files, but to no avail. I've been trying to use _collect_formant_data_fromfiles.praat and _draw_f0_curves_fromfiles.praat scripts for it.
For me, I am trying to solve an issue on why the model doesn't want to empathize on certain intonations, whereas websites like ElevenLabs seems to not have this issue (But getting the exact intonation is random).
Honestly, I have no idea if the F0 Curve would solve the desired intonation from a given audio file to convert. Please correct me if you know anything better.