For exercise 3: text-to-speech uploaded to hub as text-to-audio
#3
by
mmcgovern574
- opened
Hi, I am using the following kwargs but the model when uploaded shows as text-to-audio:
kwargs = {
"dataset_tags": "facebook/voxpopuli",
"finetuned_from": "microsoft/speecht5_tts",
"tasks": "text-to-speech",
}
I think this is the reason this assignment is marked as incorrect, any suggestions?
click on "Edit model card" on the top right corner of the model card, in the "pipeline tags" drop-down select "text to speech", then commit the changes. This should solve it.
Thanks, that fixed it!