You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Feb 26, 2021. It is now read-only.
The initiative that Mozilla has started is interesting, but I am a bit troubled that this requires "cloud" services. I think Mozilla's initiative is especially interesting if it's offline and can operate locally, with the same performance as if it was running in a "cloud". I am personally not at ease sending my voice over the Internet even with promises of it not being saved.
I am curious what are blockers to this and propose tracking them here.
Probably those could be around:
Local compute performance limitations
Local data storage size limitations
...
How does this exactly work on the "cloud" side? What would it take moving the "cloud" side entirely on the client side?
I would imagine it's possible to train a model using "cloud" resources and later execute that model locally.
Thank you!
The text was updated successfully, but these errors were encountered:
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Hello!
The initiative that Mozilla has started is interesting, but I am a bit troubled that this requires "cloud" services. I think Mozilla's initiative is especially interesting if it's offline and can operate locally, with the same performance as if it was running in a "cloud". I am personally not at ease sending my voice over the Internet even with promises of it not being saved.
I am curious what are blockers to this and propose tracking them here.
Probably those could be around:
How does this exactly work on the "cloud" side? What would it take moving the "cloud" side entirely on the client side?
I would imagine it's possible to train a model using "cloud" resources and later execute that model locally.
Thank you!
The text was updated successfully, but these errors were encountered: